The Chelate Compounds of Plutonium Frederick John Wolter Iowa State College

Iowa State University Capstones, Theses and Retrospective Theses and Dissertations Dissertations

1946 The chelate compounds of plutonium Frederick John Wolter Iowa State College

Follow this and additional works at: https://lib.dr.iastate.edu/rtd Part of the Physical Chemistry Commons

Recommended Citation Wolter, Frederick John, "The chelate compounds of plutonium " (1946). Retrospective Theses and Dissertations. 14179. https://lib.dr.iastate.edu/rtd/14179

This Dissertation is brought to you for free and open access by the Iowa State University Capstones, Theses and Dissertations at Iowa State University Digital Repository. It has been accepted for inclusion in Retrospective Theses and Dissertations by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected]. NOTE TO USERS

This reproduction is the best copy available. UMI

ras CHSIJ^TE COMPOUNDS OF PLUTONIUM , -3*- by

Trederiok Jobn Wolter

A ^esis Sttlmitted to the Graduate Faeultj for the Degree of DOOTOfi OF PHILOSOPHY

Ifajor Subjeot: Phyeieal Ghoaietry

ApproTed:

Signature was redacted for privacy. In Gliarge Major Woric

Signature was redacted for privacy. Head of Major Departa^t

Signature was redacted for privacy. Dean of ^^oate College

Zowa State &>Xlege 1946 UMI Number: DP13061

INFORMATION TO USERS

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleed-through, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

UMI

UMI Microform DP13061

unauthorized copying under Title 17, United States Code.

ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346 , -1 f;

^ I ^ ' ' TABLE OF CONTENTS

Page I IMTRODUGTIOH 1 1.1 History of the Plutonium Problem 1 1.2 The Chemioal Properties of Plutoniisa .... 3 1.2.1 Oxidation etatea ..... 4 1.2*2 Oxidation and reduction of plutoniua . 6 1.2.3 lonie species and complex formation . 10 1.2.4 The position of plutonium in the periodic table 12 1.3 Statement of the Probleoa 15 II THE NATURE OF C3HELATfi C0MP0T3NDB le 2.1 Homenolature 16 2.2 Properties of Chelate Compounds 1? 2.3 Factors InTOlTed in the Formation of Chelate Compounds 19 2.3.1 General conditions for complex for- laatioa 19 2.3.2 The cbemical bonds formed in com plex compounds 20 2.3.3 Functional groups in chelating rea gents 21 2.3.4 l&e nature of the donor atom 22 2.3.5 The nature of the acceptor atom ... 24 2.3.6 Ring formation 26 2.3.7 The possible oomplexing properties of plutoniiM 28 ill

Pag« in ISXPERKSOTAL METHODS ARI3 TSOHMlQimS Z9 3,1 Extractions on the Tracer Scale .... 29 3.a Extractions on the Micro Scale 30 3.3 Radioactive Assays . 3S 3.3.1 Plutonium assays . 32 3.3.2 Other radioactive assays .... 34 IT MATERlikLS USID 36 4.1 Flutoniim 36 4.2 Other Radioactive Isotopes ...... 36 4.3 Solvents ...... 36 4.4 Organic Reagents 37 r THE GHSLATE COKPODNDS OF PLtJTONlUM(IT) 40 5.1 Exploratory Sxperimente on the Tracer Scale ...... 40 5.1.1 Introduction ...... 40 5.1.2 Experiments with bidentate rea gents 40 5.1.3 Experiments with quadridentate reagents ...... 45 5.1.4 Discussion 51 5.2 Extended Studies v.'ith Disal ...... 55 5.2.1 Conditions for preparing the Pa(17)-disal complex on the Micro scale 55 5.2.2 The composition of the Pu(IV)- disal complex 57 5.2.3 The use of various solvents for disal extractions ...... 59 IT

Pag® 5.2.4 Behavior of the Pa{I?)-disal oomplex toward dilute aoids .... 60 5.2.5 Interfering substanoes , 62 5.2.6 Discussion . 64 5.3 The Plutonium(Hr)-Ferron Complex 69 5.3.1 Introduotion . . 67 5.3.2 ipcperimental 69 5.3.3 Disoussion 76 n THE CBLSLATl COMPOUNDS OF PLUTONIUM(III) 78 6.1 Introduotion 78 6.2 Experimental ...... 78

6*2.1 General procedure ...... 78 6.2.2 The behavior of plutonimdll} with hydroxamio aoids 80 6.2.3 The behavior of plutonium(III} with amide oximes 85 6.2.4 ^e behavior of plutoniuia{IlI) with misoellaneous bideatate rea gents .... 87 6.3 Discussion 89 711 THB USI OF OHGAKIC REAGENTS FOR THE DICONTMIKA- TION AND PURIFICATION OF PLUTONIUM 95 7.1 Introduotion 93

7.2 Sxperimental . 94

7.2.1 The behavior of various cations with disal . 94

7.2.2 lit tempts to adapt disal extrac tions to the separation of Plu tonium from zirconium and thorium . 98 •

Page 7.2.3 The behavior of various cations with hydroxamlo aoids 100 7.2.4 The behavior of various cations with amide oximes 103 7.3 Discussion 105 VIII SIMIARY 109 IX LITIRATURE OITED 113 X AGKNOWLlDGmENTS 119 -1-

I IHTRODUOTION

1.1 History of the Plutoniiim Problem

The first transuranium element was discovered in May, 1940 by McMillan and Abelson (1), who found a radioactivity with a 2,3 day half-life formed by irradiation of uranium with neutrons, and emitting negative./-rays. After showing that it differed from uranium and the fission products they identified this activity as the daughter of 23 minuter-ac ^39 tive which is formed by the radiative capture of neu trons by The new element therefore has an atomic num ber one greater than that of uranium and is elem^t 93. Their work on the tracer soale showed that element 93 has at least two oxidation states, and that oxidation to the higher state requires more vigorous conditions than the cor responding oxidation of uranium. The new element was giv^ the name neptu.iiur.. The next transuranium eleiaent discovered was elaaent 94. Late in 1940, Seaborg, McMillan, Wahl, and Kennedy (2) identified as element 94 theX-active element formed by the decay of the intermediate neptunium produced by a d, 2a re action on 0®®®, -2-

ggTr^®®{d,2n)gglip®^ 94238

Slement 94 was aamed plutoaium. ^fae isotope is aa ^-aeitter with a half-life of about 115 years. Early experime&ts with tracer ajoouats of showed that the eloaaent had at least two oxidatioa states» aad that Ibr producing the higher oxidatioa state erea more Tigorous oxi dizing agents were required thaa for the oorrespoading op

1.8 The Ghemical Properties of Plutonium

My reasonably complete discussion of the chemical properties of plutonium would require volumes. The pur pose of this brief discussion is to summarize a few of the more important properties of plutonium as they apply to a study of the chelate compounds of plutonium. So attempt will be made to aafce specific references to the original project reports In whioh the work was desoribed, as this has been done in the report of Thomas and Warner (6) and will be ooTered oompletely in the forthoomiag volumes of the Plutonium Project Record (9}«

1.2.1 Oxidation states.

ffee diposltlTe state. Of the compounds of Pu(II), PuO is the best knotvn. It is readily soluble in dilute HOI with the liberation of hydrogen, so apparently Pu{II) is not stable in acid solution. The dipositive state is produced by aetallic reduction of solid compounds of higher oxidation states* The tripositiTe state. Plutonium(IH) in aqueous solution has a blue to violet color. The absorption spec trum of Pu{III) solutions has several absorption bands, some of iirhioh are narrow. They are useful in identifying the oxidation state of plutonium in solution. The narrow bands are similar to those found for elements containing more than one 4f electron. The narrow absorption bands of the rare earths, for example, ere due to electrons rearrang ing themselves in deep-lying subshells (4f). These transi tions are thus pTOtected from the ions* environment. The general aqueous solution chemistry of PuClII) is quite similar to that of the rare earth ions. In some com pounds, Pu{III) has an ionic radius about like that of PrClII). The tetraixasitive state, me color of aqueous solu- tloas of TuilW) Tarles from green to brown, depending on t&.e anions present. The expleinatlon of this In tems of oompXex Ion formation will be dlsoassed later. I'he absorp tion peaks of Pa(17) are sharp and narrow, and serre as a means of Identifying Pu(IV) in aqueous solution. !Bie oh^lstry of Pu^IT) is very similar to that of Ge(I7) and Th(I^), ezoept in its ozidation<»reduetion propertieSii Zt la muoh more stable than tJ(XT), and its ion- ie radius Is sll^tly less than that of Ce^IT). The hydrox ide of Pu(r?) is amorphous and extremely insoluble. It shows no amphoteric properties. This hydroxide has a ten dency to fom a green solution oontalnlng rarlous sized ag gregates, some of whioh have been found to be of oolloidal dimensions. The hydroxides of Th(IV}, U{17) and Ce{17} al so exhibit this same phenomenon of dispersion into oolloidal partieles. ^Qie poljpmerle form of the hydroxide of Pa(IT) requires heating in concentrated aold for solution, and in dilute aoids the transformation from oolloidal to lonie foxm is often Tery slow. In eertain media Pudv) is the easiest of the rerious oxidation states to maintain. The pentapositlve state. Only one compound of PuCT), believed to be potassium plutonlte, has been described, al- thou^ there is some evidence of PuFg compound. Pu(T) has been detected spectrophotometrioally as an intermediate state in the reduction of Pa(7I) with hydroxylammonium ion, SOg, and Eg. Apparently T?m{7) is unstable with respeet to Pudll), Pu{I?) and Pu(TI).

file ]a.exapositiYe state. The eompouade of Pu(7I) and their aqueous solutions exhibit eolors from pinlc through orange to brown, and the aqueous solutions have the oharaoo teristio sharp absorption bands in their spectra. fhe aqueous ohemistry of PuC?I) is very similar to that of and the plutonyl and plutonate salts resmble Tery elosely the corresponding uranyl and uranate compounds. The ionic radius of Pu(YI) is only slightly smaller than that of U(TI), and Ptt(VI) exists as PuOg** Ion in the ab sence of complexing anions. There is no fluorescence corresponding to that exhibited by the ioa. Higher oxidation states. There have been various at- t^pts to prepare plutonium of oxidation number greater than six, but they have been unsuccessful.

1.8.2 Oxidation and reduction (8f plutoniuM

yormal potentials lietweea Pu{m). PuflY) and Pu{Yl). detezmination of the potentials between various oxida tion states of Plutonium was one of the important objectives in the research program on the ohemistry of plutonium. ®bie various potentials listed below are the best values obtained from direct cell measurements, polarographic deter minations, and spectrophotometric measurements of disproportionation equilibria. All values are in 1 M acid at 25^C. -7

HCX PmClII) -O.Wi Pu(IV) -1.049^ Pu(7I) , -1.019 ,

BMO3 PuClII) -0,924 Pu(IT) -1.114 Pu(7I) , ^-1.042

HOIO^ PuClII) -0.95 Pu(IV) -1.067 Pu(TI) , 1.027

H2SO4 Pu{m) -0.75 Pa(I7) -1.2 to -1.4 Pa(71)

The Fa{IT)-Pu(71) eouple 1q 1 M HgSO^ is estimated from measureanents of irreTerslble eells (10). There have been no direet measurements in neutral and alkaline solutions, in whioh the relationships are muoh sore oomplioated. goraal potentials of the Pu{III)-Pu(iy) oouple. flie potentials of the Ptt(III)-Pu{IT} couple have been measured under many conditions if^ioh are of general interest in the chemistry of plutonlvua. Reference to Table 1 will show the eTidence of ecMplexing of Pu(I7) by various anions (10, 11). Table 1 Toimal Potentials of the Pu(III)-Pu{IT) Couple in Tarious Media at 25<^C

Soltttion 1 M HAc, 1 H HaAe -0.396 0.3 M mag, 0.1 M NaAc, 0.05 M fia2S04 -0.261 1 M HGl, 1 M HF «0.50 1 M HOI, 0.1 %P04 -0.80 1 M BOX, 0.6 M SIFOI -0.59 0.5 M H£S04, 0.6 M %P04 -0.60 0.05 M I2&O4, 9 U IiH4F L-0.87 O.OS M B2SO4, 0.18 M &&G2PA, 0.18 M ££804 -0.234 0.05 M H20g04, 0.75 U Kz%Qa, 0.05 M E2SO4 -0.082

Disproportionation equilibria. In solutions in î^ there are no anions "îch tvill stabilize one oxidation state of Plutonium throxigh complex forzoation, disproportionation aông the Tarious oxidation states occurs, as a oontequenoe of the fact that the potentials of the Pu(III)-Pu{lT)-Pu(VI) couples are so nearly equal in ralue. The existence of these equilibria greatly complicates the aqueous solution chemistry of Plutonium. ®ie PuCl?) ion is relatiTely unstable. In dilute solu tions of HSl, HHOg or HOIO4 it undergoes partial disproportionation into Pu(III) and Pu(VI). In 0.5 M HOI, the Talue of the equilibriuiB constant (in molar concentrations} is 01052 at rlHtea tempernture for the reaction -9-

• ZSzO S« Ptt*4 f sHgO PuOg*'^ k 4S^ ^ 29

IM fi2S04 solutions and in otbsr media in whioh Pa(lY} is stabilize by oomplex formation the equilibrium constant for the disproportionatioa reaotion beecaies very small. The equilibria in the reaetioa

Ptt(T) • Ptt(IT) Pu{III) • Pu^TI) hare not been studied so thoroughly, ai^ values of the equili- brim constant range from 5.4 to 8.8. Oxidation-reduot ion reaotions. Referenoe to the formal potentials of eouples of the varioius oxidation states of plu- tonitsK ivill suggest ti^at media might be necessary for Tarious oxidation reduotion reaotions. for the oxidation of Pu(III} to Fu(I7), oxygen in the presenee of HgSO^ is effective, as are concentrated HHO^, Olg, and ©PgOy*"" with HHOg. For the oxidation to PuCYI), so dium bismuthate, BrOj", Cr^Oi^ , Ag***" and SgOg » 01004, Ce**'*'*'» and electrolytic oxidation hare been used. for the reduction of Pu(7I) to Pa(Z7), fuming with H^SO^ is particularly effective, for the reduction of Pu(IV") to Pii(III) SHgQH*, Hg. aad SO2 have been used most extensively. Other reductants which have had wide application in various operations are Fe*^, I", H£0£, formic acid, and foxmaldehyde. .10.

St&t>iXizat>ion la emy parti oular oxidatioa stato i» often determined by the presenee of oomplezing anions. The Icinetios of ozidation-redaotion reeotions hare been studied only rather superfioially.

1»3»3 lonio speoies and oomplex formation

Tarious methods, suoh as S. M. ?. determinations, solu bility measttriuaents, absorption speetra, and eleetrieal trans- fereaoe experiments, have been used to determine the nature of the ions of plutonium which exist in aqueous solution. lonie speoies and eomplexes of PuClIlK The absorption speotra of Fu(III} in 1 M solutions of HCl, HCXO4 are oXoseXy simiXar, suggesting that the same ionio spe oies, probabiy hydreted Pu*'*'*', is present. However, tiiere is 8<»ae eTidenoe from ehemioaX studies and eXeetrioaX migration data for possibXe oompXexing of Pu{III) by 1. H. ?. measoreaeats on the Pu{2II}-Pu(X7) ooupXe in HgS04 do not shed any Xight on this, beoause of the Tery strong oompXexing of Pu(I7) by suXfate ion. IridwELoe of oompXexing by oxaXate ion is obserred in the soXution of the insoXubXe Pu{III} oxaXate in saturated %Og04 soXution. lonie speoies and oompXexes of Pu{17). Absorption speo tra and E. M. F. data have reveaXed no evidenoe of eompXex- ing of PuCIV) by HOX and HCXO4. Sitrate ion oompXexes Pu(X7j quite strongXy. The exist- -11 teaoe of PulMOg)*'**' and PuCHOgJg**' ioos h&B been shown by absorption speotra measur^ents, and the complex anion analogous to hezanitratooerate, has been found from eleotrieal migration experiments to exist in 8 to 10 M HID3. 1. M. F. measurements have indicated the oomplexing ae«> tion of SO4**, and absorption speotra deteriainations have been interpreted by assuming the existenee of oomplex sulfated eations. In 0.1 to l.OM HgSO^ solutions, eleotrieal migra tion data have shown that Ptt(IT) exists predominantly as an anion. fhe foraation of oxalate eomplexes of Pu(IY} has been shoim by solubility studies of the oxalate in Tarious oxalie aoid oonoentrations. Similarly» there is evidence of a oi- trate eoiaplex in the observation that PuClOg)^ dissolves in sodium eitrate solution. •Ehe aoetate oomplex of Pu{17) is apparently quite stable ^ preventing the precipitation of the hydroxide at pH*s as hi^ as 5, as compared to precipitations at pH's below S in the ab- senee of aoetate. In 1 M aoetate solutions at pH 3.5, Ftt{17} exists as an anion, and when a Fu{17) solution is treated with exeess sodim acetate, the color changes from green or brown to a purplish orange. Ionic species and ccaaplexes of Pu(7I). Ti»re is no speotrophotometric evidence for complex formation by PuCTI) in either H8O3 or HCIO4 at 1 M concentrations. However, In -12-

9.4 M 8oae oomplex ion formation does oooux. There is some oomplex ion formation in 1 M M.SO. and in HCl of oon- 2 4 eentrations as low as 0.5 M. By analogy with UOg*"* it was proposed that Pu(TI) would probably exist as The Pu(in)-Pu{?I) and Pu{IT)-Pm(VI) oouples haTe a fourth power hydrogen ion dependenoe, in aeoord- anse with the reactions discussed in the section on disproportionation equilibria (1.2.2}.

The existenee of PuDg**" is supported by a great mass of ohesiiaal and physical eTidenoe showing its similarity to In ooatrast to Pu(IT}, FuOg^** does not precipitate as the hydroxide when its solutions are made basic. Plutonates or polyplutonate double salts appear when PuOg*** solutions are made basic with HaOH. 13iey precipitate more slowly and are more soluble than the corresponding compounds. ^e presence of a holding oxidant is usually necessary to maintain plutonim in the hexafalent state.

1.8.4 ^e position of Plutonium in the periodic table

Before the discovery of the transuranium elements there was considerable speculation about the possible electTOnic con figurations of the clients beyond uraniisa (IS). The fact that not a great deal was Icnown about the chemical properties of the elements immediately preceding uraniim in the periodic table added to the uncertainty. After the discovery of the transuranium elements through 13 elê&t 96, enough o^mioal Informatioa about the heavy ele ments was aeoiKiEiulated to make possible oertaia deduotloas about the atomic struotures of the elements in this region of the periodic table. It had been thou^t possible that a transition group should appear in the neighborhood of these elements. If a 64 transition group (siisilar to the 5d group from hafnium to gold and tile 4d group from yttrium to sllTer) were being fomed, neptunium i«ûld be expected to res^ble rhenium, and plutoniua might resemble osmium. However, neptunium and plutonium are mueh more electropositive than rhenium and osmium. There is no evidence for a volatile FuO^, in contrast with volatile os mium and ruthenium oxides, and the ozidation state of eight has never been described for plutonium. Thus it seems likely that the transition in the elements from 69 to 96 does not in volve the simple filling of 6d orbitals. There is considerable evidence that in this transition group the 5f shell is being filled, and that the series begins with actinium in the same sense that the rare earth series be gins with lanthanum (7). It has been called the 'âctinide'* series, by analogy with the ^'lanthanide" series for the rare earths. 5ê valence states of the elements from 89 through 96 can be explained in the same manner as the valence states of the rare earths (13). Thus it wuld be expected that ele ments 95 and 96 should exhibit very stable tripositive states. -14 and that 9fi should exist almost exclusively in the triposi- tiTe state, because with its seven 5f electrons, its elee- troaie structure should be aaalogous to that of gadoli&iTiffl, which has sevea 4f electrons. Chaaical studies have ôwa that the tripositive state is characteristic for element 96. On this basis the names proposed for elements 95 and 96 are "asiericiua," Am. {by analogy with europium) and "curium," Cm {by analogy with gadolinium} {14}. Other evidence substantiating the hypothesis of an "ac- tinide** series includes meesortH&ents of the magnetic suscep tibility of uranium and plutoniua, the sharpness of the ab sorption bands in the spectra of aqueous solutions of these ions, end evidence of a coordination nisiiber of e^ht for IT{17) and Fa{IY). The relationships may not be so sharply defined as in the "lanthanide" series, because the energy difference betwera the §f and fid shells is probably very small, and changes in environment and state of chemical combination might determine the positions of the electrons with respect to the 5f and &d orbitals. Bie eELission spectrum of plutonium is too complex to be of any value at present in determining the electronic config- uratioQ. of Plutonium. -15.

1.3 Statement of tlie Frobl«ft

Xn oonaeotlon witii studies on the general oboaistry of plutoniim, a pro&e&m. was authorized for a study of the (Oie- late oompounds of plutonium eztractable Into organic solvents. Zf suitable organio reagents speoifio for plutonium oould be fouM, they might prove of value for such operations as ex traction, decontamination, oono«itration or purification of plutonium. Decontamination is the process of elimination of the hi^J.y radioactive fission products which are formed along with Plutonium in the chain reacting pile. The chemical se paration of gram amounts of plutonium from ton amounts of uranium is a complex problem, as the requirements for the de contamination and purification of plutonium from troublesome el^ents are very rigid* ^is thesis contains part of the results of the studies designed to find such organio reagents for plutonium. 16

H THE SATDRE OF CHELATE OCMPOmiDS

2*1 Momenelatar•

flie term "eh«late", proposed by Morgas aad Drew {15} to designate those oyolio struotures fomed by the imion of aetallio atoms with organio and inorganio molecules, is de rived from the Greek chela« which refers to the great claw of the lobster. The term is applicable to these ring sys- tffi&s since the associating molecules are characterized by a pincer-like structure. More generally, chelate rings refer to cyclic structures which arise through intramolecular coordination in systems containing a donor and acceptor center or through intermole- ciilar coordination in systems capable of forming two or EJore coordinate links. Sxamples of intrasioleoular coordination are salicylalde- hyde (1} and copper glycine {II)s

.0. .0, 0 s C C s 0 ' \ ! -H CHg / MHg HjR I n -17-

Sxesaples of the iatexmoleoular type eore the dimers of earbozylio aaids (III) and the Werner complexes deriTeS from ethylenediamine (lY)t

0-H 0 w / H-C 0-H ^2 Au Br \ / GH2 0 H-O \ R NH Z in IT

The oyalio struotures may be formed by primary valences (those in which hydrogen is replaced by metal), by secondary or coordination valences, or by combinations of the tvo, Re agents capable of forming cyclic complex structures hare been classified as bidentate ("two-toothed")» tridentate, quadriden- trate, and so on, in which the functional groups may be aci dic, coordinating, or both. Diehl (17) has published an ez- eelleat survey of chelate rings based on the above system of classification.

2.2 Properties of Chelate Oompounds

•l&e non-electrolyte chelate compounds, those in which all of the primary and secondary valences are satisfied to produce a complex with a zero net charge, are of special interest in 16 analytleal ehemlstry. These are known as laaer oomplexes. Suoh oom|K>aadfi are usually highly oolored, aiê insoluble la water and other polar solvents, and are quite soluble in nonpolar organio solvents. IQiese solubility properties lead to unique uses in analytioal chexaistry. Thus, if an organio re- ag^t is spec if io in forming a true inner complex of the ele*> ment in question, there are several advantages of extraction methods over precipitation procedures. Solvent extraction of inner complexes can effect much greater separation of a major element from minor amounts of impurities than is possible by precipitation methods because two->pha8e extraction separates elements without the presence of a solid phase capable of ad sorbing ot occluding undesired ions. These factors maice sol vent extraction of inner complexes particularly applicable to the separation, without carriers, of minor components from large amounts of other metals not complexed by the chelating reagent. Actually, the search for a specific reagent is a rather ambitious plan* Although a great deal of work has been done in tê preparation of chelate compounds and upon their strueture» there apparently is still no method for predicting the specificity of a reagent for a particular metal ion. We may draw upon many generalizations that havie accumulated first from the vast amoimt of classical work on ^''emer complexes and more recently from develojênts in sttidies on the nature -19- of the chmloal bond In complex eompounde. However, the search for laore or less specific reagents still mast be conducted in the laboratory, using as a guide certain rules based on experimental results obtained in ata- dies on all kinds of complex compounds. One author has been led to state that **18 extrapolating past experience—we must be careful not to leave the laboratory long to theorize on what we think should happen with an untried reagenf* (16}.

2.3 Factors Involved in the Formation of Chelate Compounds g.5.1 General conditions for complex formation

Complex formation depends on the nature of the acceptor and donor atoms and the types of bonds which may be formed between them. The stereochemical factors are of isrime im portance, too. The subtle relationships among these factors are too complex to permit predictions of quantitative behavior, so the usual approach is to study the variables separately. In addition to the structural relationships mentioned above, there are also peculiar effects of hydrogen ion concen tration which determine the specificity of reagents for cer tain cations. For example, diphenylthiocarbazone forms chloroform*>8oluble complexes with many metals, but by controlling the hydrogen ion concentration of the medium in which complex formation takes place, tthe specificity of the reagent can be improved {ref. I9a, p, 91 ). OupferroB and 8-hydroxy- quinollBe are also exaciplss of rather rabid oomplex-formlag reagents ii?hose speoifioity can be iaprored by proper oontrol of hydrogen ion oonoentration. There are sereral reference books on the use of organio reagents for metallio ions (19) which might serve as sourea material for a study of the problem of finding new reagents for metallic ions» The work of Sidgwick: (20,21) and of Pfei- ffer (22,£3,24) has opened the way for the syst^atic searoh for new organio reagents for metfcllio ions,.

8»5»,2 fhe oheaical bonds formed in eoaplea: compouads

Ihe binding forces in complex compounds may be due to electrostatic attraction for the surrounding ions or oriented dipoles^ or to true covalent binding, or to some combination of these types., Ffeiffer {ZZ>) contends that in inner implex oofflpounds the primary and secondary valence bonds differ only in the origin of the binding electron pair.

la FedgO)®^^'', Hi(H^)/*'. aad many similar ooapl«c ions, the bonds between the central atoms and the surrounding molecules result in a large part from ion- dipole attraction* llectrostatic bonds might result frcaa the attraction of an ion for the induced dipole of a polarizable molecule. Govalent bonds involve the sharing of electron-pairs be tween two atoms. Such bonds may be formed by each atom fur- ••2X— aishlng one eXeotron or by both eleotrons being furnished by one atom {the donor} to fill a raeant orbital of the aceeptor atom. In most oases, the bonds are formed bj '^hybridization,*' whieh is a oombination of s and p and other orbitals. The applloation of isagnetio measurements to the study of ooiilplex compounds is due largely to Pauling (25) and most of the re cent literature on the subject is expressed in terms of atomic orbitals and other eharacteristies of Pauling*s nomenclature. Umball (26) has summarized the possible bond types for rarious coordination numbers, giving electronic configurations, geometrical arrangonents and relative bond strengths* Further discussion of these factors t?ill be considered in Section 2.3.5.

2.5.5 Functional groups in chelating agents

In the formation of non-electrolyte complexes two kinds of reactive groups are involved, acidic groups and coordinating groups. Among the various functional groups which may unite with metals by primary valence by replacing hydrogen by metal are the following:

earboxylic acid - COOH sulfonic acid - SO^fi eaolic or phenolic hydroxyl - C-OH oxime-nitrone - * NOH primary amino - -Mg secondary amino - s IS -22-

fhe seoondary Taleaoe groups, whloh oontaln so-ealled doaor atoms with aa unshared pair of eleetrons, form ooor« dinate honds by the donation of this unshared pair into ra- oant orbitals of the metal ion to form homopolar bonds. These donor atoms are contained in suoh groups as the following:

primary amino - -HHg secondary amino •> sM tertiary amino - -K cyclic tertiary amino - H / imino - «lf- oxime - shoe alcoholic hydroxyl - OH carbonyl - 0=0 thioether - -S>

fhese lists could be expanded, but with little point, as most of the known chelate rings involTe combinations of the functional groups giren above. Most of the organic reagents known and commonly used are of t^e bidentate type, containing one acidic and one co ordinating (or secondary faience) group.

8.S.4 The nature of the donor atom

In complexes containing the coordinate structure A^BO, in which B is the donor at(»ti, the strength of the coordinate 23-

boad oan seldom be detem^ned* AS a rule vse oan disoOTer the affinity of particular donor atoms for particular accep tor atoms only in a qualitative way trom. a consideration of the ntacber and stability of the complexes fomed by the tiwo kinds of atoms.

I^e nature of a substituent (C) on the donor atom may modify the strength of the link. This has been noted in the change of stability in the ammines of Cu(II), Co(II} and Hi(II) when the hydrogen of HHg is replaced by hydro carbon radicals (SI), and in the base strengths of various amines. In many organic compounds the effect of substituents on the donor atom appears to be due to the polarization from the inductive effect (27) proposed by Ingold to explain the effects of substituents on reactivities of organic com pounds. Orleman (26) has used the idea of the inductive ef fect to explain the coordination effects noted in the extrac tion of Plutonium nitrate from aqueous solutions by various ethers and ketones. fhe stability of the A-^-B link in the A-^BC complex al so depends on the nature of the groups attached to A by aci dic linkages. Spacu and Voichescu (29), in studying the va por pressures of Nl^ over aiomines of various cupric salts, found that the strength of tdie Ou-trllHg bond was inversely proi>ortional to the base strength of the anion of the salt under otherwise comparable conditions. Oalvin and Wilson (30) £4- bare reported that in the ouprio chelate oompounds vrith sub stituted salioylaldehydes and y^-diketones there is a direot relationship between the stability of the complex and the acid strength of the reagent. The generalizations obserYed in the study of non~ohelate complex compounds can be used as an approach in the study of chelate compounds, but only to a limited extent. This is true because in the polydentate organic reagents both acidic and co ordinating groups must be in the same molecule. Tor example, the same substituent groups that increase acidity will decrease the donor power of a donor group. In en organic reagent con taining both acidic end donor groups the proper balance between these two opposing effects must be attained. a.3.5 Ighe nature of the acceptor atom

Long before the electronic configurations of the ele ments had been deterxnined, it had been observed from experi mental facts that the transition elements are most easily in troduced into coiiplex compounde, even though many other ele ments also display complex-forming properties. Sidgwick (20,21) attempted to explain complex formation with the idea of the "effectire atonic nimber" {1. A, H. ) of an element, vîch is the total number of electrons either pos sessed completely by the atom or shared with other atoms in coTalent or coordinate bonds. Sidgwick postulated that the 25. teadeno^ for oomplex formation could be explained by assum- lug that an ion tended to achieve the 1. A. N. of the next inert gas hj gaining electrons from a donor atom. Thus he explained the formation of NiCCO}^ on the basis that ggHi, with 1. A. N. « ^ added 4 pairs of electrons from CO mole. cules to form Ni(OO)^ with S. A. N. - 36, the E, A. H. of Icrypton* While Sidgwiok*8 idea would explain the formation of many oomplex compounds, it had many shortcomings, espe. cially in that it could not explain the valences of oomplex ions of the transitioh metals. In the extremely stable acetyl, acetonates of Mg, Zn, Al, and Th the £• A. H. is that of an Inert gas, but with Ni and ?e the acetylacetonates are stable and seem to exist without respect to the £. A. N. After more was known about the electronic structure of the elements and after the quantum mechanical approach put the nature of the covalent bond on a somewhat more rigorotia basis (25), it was possible to explain differences in the oomplexing properties of various ions (31,32)• The mcêra explanation for the coaiplex-forming properties of the transition elements is that they have low-lying vacant orbitals which can form homopolar bonds by accepting electron pairs. The concept of hybrid bonds has been substantiated by magnetic measurements, stereochemical evidence, and chemical properties. Th\xa in Zn(CH)^ the sp® bonds give a tetra. hedral configuration, Ni(0N)4 — YioJuSiB square. -26- and Fd(Gl)0 "" with dâp® bonds is octahedral. Selifeood (33) has disoussed complex ooapounds quite tho roughly with respect to the use of magnetic measur«rieats in showing what orbitals are involved in the actual structure of GQDiplexes. In complexes in which all of the bonds are not covalent but in which some of the bonds may be partly ionic or elec trostatic, it has been found that the coordinating tendency of the metal ions increases with the charge density on the eentral atom. That is, the bonds are stronger the greater the charge and the less the ionic radius of the eentral atom.

8.3.0 Ring fommtioa

The possibility of closed rings in coordination compounds was first suggested by Tschugaeff in 1907 (34) to explain the behavior of metal impounds with biuret and similar substanoes. Werner (35) developed the idea imiaediately afterwards, and since then siany generalizations have been used to explain the stability of chelate compounds. Eing formation generally increases the stability of com plexes. For example, ethylenediamine forms more stable com plexes than primary amines. Phenol is a weak complexing agent, yet catechol forms complexes with many ions. Oxalate coffii- plexes are far more numerous and more stable tJtian complexes with monocarboxylic acids. Xf there are no resonance possibilities, the most stable -27- rings are those of five or six members, as might he ^peoted from strain theory. However, chelate rings with as maBiy as thirteen members have heen described (26). Pfeiffer (23) and Sidgviek (20) have classified the most common ring types ia chelate compoands in which various combinations of G, IS, 0 and metal atoms are considered. In structures containing possibilities for resonance, ring stability is greatly increased. Thus chelate rings, formed trm. ^-diketones, ^-keto esters and ccmpouMs in which the ring includes part of an aromatic structure my be stabi lized by resonance. A classic example of stabilization through resonance is copper phthalocyanine, which is so stable that it resists boiling HCl and molten caustic, and can be sublimed un changed at 580*^G at reduced pressure in nitrogen atmosphere (37).

\/^0

It has been suggested (36) in the case of nickel dimethyl- glyozime that the electronic structure of the divalrat nickel is aa essential faotor in establishing the resonance of the whole complex. Similarly, it has been proposed that organic -S8- reageata which are specific for particular ione owe their apeoifieity to the electronic arrangexaent in the central atom and the partioipation of the central atrasi in the resonance of the whole complex.

2.3«y The possible complexing properties of pliitonima

If chemical experience had shown that plutonium resembles omivm and ruthenium (the transition elements in which 5d and 4d orbitals, respectivelyt are being filled) we might ex pect plutoniim to be an active complex former. However, it ap pears quite certain that the transition group of which pluto nium is a m^ber is formed by the filling of Sf orbitals. Therefore, it is not expected that plutonium should be uniquely active in complex formation. The principal approach, then, %ould be to determine the coordination number of Pu(lII) and Pu(X7) by the luse of reagents icnown to complco: many ele ments. Then, knowing the coordination number assooiated with eaeh valence state, the polydentate reagents tried should be those which could lead to non-eleotrolyte complexes soluble in nonpolar solvents. Ihile Pu{VI) might form inner complexes, the necessity of using "holding oxidants^ to maintain the hexapositive state makes it impossible to use some organic reagents which are sensitive to oxidation. 29-

III MPEHIMlliTAL METHODS AHD TECHHI^UIS

3.1 Sctraotions on the Traeer Seale

In all of the exploratory experiments with traoer plu- tonlom, the experimental prooednre was essentially the same. The aqueous solution oontaining traoer was buffered to the desired pH with sodium or ammoniim acetate and aoetio aoid in solutions 0.1 to 1.0 H in total aoetate, Hlhoi only ap proximate values of pH were desired, indleator paper drion** paper, manufaotured by Mioro Sssential Laboratory) was used. For more preoise values, glass eleotrode pH meters (Beeicman Model Q, Leeds and l^orthrup Model 7661-Al) were used. The usual solvent was ohloroform, since it has rather exoeptional solvent properties for most true inner oomplexes. In oases in whieh the organio reagent was relatively soluble in water, the reagent was added to the buffered traeer solu tion. When the reagent was insoluble in water, it was eon- tained in the ohloroform. In all of the exploratory experiments reported, the volixme of the solvent layer was equal to that of the aqueous solution. The volumes were usually from 10 to 15 ml, and the amounts of reagent were from 10 to 20 mg. The extraotions were tarried out by agitating the two phases vigorously in an ordinary separatory funnel. -30-

Aft«r separation of the phases, the aqueous phase was evaporated to a small TOlume for assay. For analysis of the organic phase» the solvent was evaporated and the or- gaoio residue was destroyed with fuming HSOs and HOIO4, and talcen up in water for assay. In sGQ^e oases, oarrier ions, known to react with the chelating reagents under investigation, were added.

3.2 Sztraotions on the Micro Scale

For operations with amounts of plutonium ranging fxoa 50 to S or 3 Bg, specially designed micro equij^ent had to he used. The pipets used were designed by R. R. Baldwin, and a representative type is shown in Figure 1. The actual pipet is of quartz (A) and is sealed into the sheath (B) with de^- ISiotinslcy cement (£). To fill the pipet the hole (C) is closed with the finger tip while gentle suction is obtained with a hypodermic syringe attached at (D). The liquid fills the entire pipet. To empty the pipet, pressure is exerts on the hypodermic syringe. The pipets were calibrated **to contain** by weighing tiie volume of mercury which they contained. In order to effect ocmplete transfer of solution, two or three washes were ne cessary. Excellent reproducibility could be obtained with pipets with capacities from 1 to 300 microliters. Pig. 1. Cross Section of Micro Pipet

Fig. 2. Micro Extraction iipparatus 32«

With amounts of plutoalua exoeediog 100 separatory funnels oould not be used for twoophase extractions. %ere- fore, a micro-extraction apparatus, adapted from a design by Langhaa |39) was used. This apparatus is shown in Figure 2. The extraction thimble (C) contained about 1 ml of the aque ous solution or suspension to be extracted by chloroform. Ohlorofoim (1 iol was introduced into the sidearm of the thimble. For agitation of the two-phase system stopcocks D and B were opened, stopcock F was closed, and an inert gas (helium or nitrogen) was introduced at E at a rate slow enough to agitate the two phases in- the extraction thimble with the iasired vigor. When extraction was complete stopcock D was closed and stopcock ? was opened. Then the inert gas was in troduced very carefully at A. The hearier solvent layer was forced up the tube and drained into the receptacle 6. The flow of gas was stopped the instant the chloroform-water in terface reached the tip of the tube at H. The water layer in C and the chloroform layer in 0 were then analyzed in the usual manner.

3.3 Hadioactive Assays

S.g.X Plutoniiaa assays

The alpha particles from and Fu^® hare ranges of

about S.5 to 4 cm of air. This corresponds to a mass range of about 4.9 ia air, or of about 10 mg/om^ for the carriers usually used. Therefore, to avoid low results due to selfoabsorptioa la the sample, rery Uiia sasiples had to be prepared for alpha oounting. For analyses on the traoer soale, the samples were pre* pared in the follovîng iuanner. The aqueous solution was treated with iÔg or SOg to produce the fluoride-insoluble Pu{III) or Pu(I7) state, tA*** ton was added as carrier and LaFg precipitations were made from solutions 1 to 5 M in HF. The fluoride precipitations were done in specially prepared Lusteroid centrifuge tubes with bottoms flattened to foim a smooth surface, fhe suspension was then centrlfuged, the supernatant poured off, the precipitate carefully washed with a fine stream of water and centrlfuged again. With proper care, a uniformly thin LaF^ precipitete could be deposited on the bottom of the lusterold tube. After the second centri- fugatlon, the supernatant was poured off, and the precipitate on the bottom of the tube carefully drl^ by a gentle stream of air while the tube was gently warmed above a hot plate. After the precipitate had dried^ the bottom of the tube was cut off with a razor blade to form a very shallow cup with walls about 1/32 to 1/lft of an inch high. This disc with its LaFg deposit of leas than 1.5 ag/wi^ was then used as the sample for alpha-counting. For assays on the micro scale no lâFjj precipitations were 34- used. Suitable aliquots of solutions were transferred with micro pipets to platinum foila and oarefully evaporated to produoe a uniformly thin deposit. The margins of the plati- mm foil were ooated with Zapon laoquer to prevent the solu tion from oreeping over the edges of the foil. Sometimes a drop of ethylene glyool was added to aid in the slow evapora tion to a uniformly thin deposit. After the solution was eva porated to dryness, the foil was heated to a dull red heat to get rid of organic matter and volatile salts, This foil» with its extremely thin deposit, was used as the sample for alpha- eounting. 13ie alpha activities, if less than 3000 counts per minute, were measured on a standard alpha counter (40), with a scale of sixty four circuit and mechanical recorder. The geometry of the counter, as deteKained hy R, H. Baldwin, was 51.656. For activities greater than 3000 counts per minute, and for samples with a high /-activity, a methane chamber proportional counter, with a scale of one hundred twenty eight circuit was used. The geometry of the proportional counter was 50^ (41).

3.5.a other radioactive assays

In experiments dealing with radioactivities of fission products and consisting ofand /^rays, samples for measure ment were prepared either by evaporation of aliquots on plati- mm foils or small watch glasses, or by precipitation with -35- earriers. Less oare had to be taken la the preparation of thin sampleB than in preparing plutonium aaraples, becatxs® of the greater penetration of the c/- and jf-rsjB, The radioactivities were measured on an electroscope or a Geiger counter depending on their strength. Ihe electro- seope used was of the Lauritsen quartz fiber type with a S.4 mg/oB^ aluminum window. The samples were measured by placing them under the ionization chamber in a holder which kept th«a in a reproducible position. The Qeiger-Muller counter was of the copper wall type with a mica window 5.B mg/cm^ in thick ness. A scale of sixty four circuit and mechanical recorder were used. -36-

If MATERIALS TISSD

4.1 Plutonium

238 The supply of Pu was isolated from uraaium metal bom barded with deuterons at the Berkeley oyelotron. 2S9 Solutions of pure Pu , isolated from pile uranium, were obtained from the Clinton Laboratories. Dilute solutions of used in traoer experiments, were obtained by proeese- ing slugs of uranium subjected to slow neutron irradiation in the Clinton pile,

4.2 Other RadioaotiTe Isotopes

In some of the studies on the specificity of organie reagents, other radioactive isotopes were used. Solutions of S4 day Th^®* were prepared by the ether extraction of uranyl nitrate. Other radioactivities, including 33 year Gs^^, S0 day «75 day 57 day 68 day Zr®® and 12.8 day Ba^^, were obtained frcaa supplies of activities which various members of the Ames project had isolated in other studies.

4.S Solvents

ehlorofoxm and carbon tetrachloride were reagent -37

obtained from Geaeral CheEaieal Company. All of the other solTents used were obtained from Eastman Kodak Company.

4.4 Organio Reagents

The following oompounds, which were used directly in examinations for complex formation, were obtained from last- man Kodak Company: quinalizarin, benzoin,^-benzoin oxime, aoetylaoetone, benzoylaoetone, dibenzoylmethane,.K-nitroso- ^/-naphthol, ^-nitroso-p^C-naphthol, dimethylglyoxime, isonitrosoaoetophenone, isatin, 8-hydroxyquinoline, 7-iodo-8- hydroxyquinoline-S-sulfonio acid, diphenylthiocarbazone, diphenylthiourea,

H-OOO-CgHg • MHgOH KOH —H-COHHOK • CgB^-OH • HgO g-CQHHOK • C%COOH -v BCOMOH f C%COOK -38-

dlthlooarbamates were prepared by tbe method of IJelephiae (45) using the general reaction:

S H « R-HHg • CSg • HH^OH • HgO

Di-substituted dithiooarbamates were prepared by a similar reaetion using a secondary amine as starting material. All amide oximee were prepared by the same general me thod, adapted from that of Tiemann and Kruger (46)s

HHg I S-CH • HHgOH'HOl • HaOH B-0=N0H • KaOl • HgO

For the experiments with the diimines of o-hydroxyaldehydes, many of the dlimines were obtained from Dr. Harrey DiehX. Zn the other oases, the diimines were prepared by re- fluxing the aldehyde with the appropriate diamine in ethanol, and the Sohlff's base crystallized from ethanol (47). For the preparation of the o-hydroxyaldehydes not direotly obtain able, the method of Duff (46) was used. The synthesis con sists essentially in causing hexamethylenetetramine and a phe nol to react in the presence of anhydrous glycerol and glycero- borie acid at a temperature of 150 to 160®C. The o-hydroxyalde- hyde in nearly pure form is obtained by steam distillation of the acidified reaction mixture. The mechaMsm suggested by mtf is -33-

m f S -H / —CS2"H®CH2

-OH -OH vrator ^^-CH2-IPC^2 -0H=H-0H3 -CHG

Other misoellaneous ocmpounds were pre|>ered by methods described la the literature* and inoluded aoetylaeetone sioao- imlde (49), salloylalpheiiyXhydrazo&e, o-hydroxyaoetophenone semloarbazone,

7 THE CHELATE COIPOUNDS OF PLUTONITJaflV)

S.l Exploratory Experiments on the Traoer Seal®

8.1.1 Introdttotloa

l&>st of the organic reagents oommonly used for metal lone are hidentate moleoules, in consequence of the fact that the coordination number is generally twice the primary Talence. Therefore, the most obTious approach to the probl^ of finding organic reagents that oomplex platonium is to make use of the more actire reagents knom to complex many ions. Frcâ the data obtained in this way it should be possible to detemine the coordination number of Pu(I7}. ên» knowing the coordination number of Pu(IT), a more systematic study of group interactions and the effect of che late ring size could be made to obtain more specifically the requirênts of organic reagents which will form chelate oompounds with Pu(lY).

5.1.g Bxperiments with bidentate reagents

All of the bidentate reagents, containing one donor and one acidic group« which were examined for complexing activity with Pu{IT) tracer were known to form true inner cx^mplexes with some other elements. In all of the tracer experimeata the procedure was essentially that described in Section 3.1. 41'

Clalorofom was used as the solvent in all of the extreotiona except in the experiiaents v.'itli quiaalizaria, in waicli aniline was used, and in those with trifluoroaeetylaoetone, in whleh benzene ms the solrent. The z-e&ults obtained in single ex- traetions with a volume of solvent equal to that of the aque ous phase are summarized in Tahle 2,

Table 2 Behavior of Traeer Plutonium(IV) with Various Bidentate Organie Reagents in Aoetate-Buffered Solutions

snKaBaBBmDKBBHetttaBmmtBaBBBHnBBBncaBMBBHBBBBBaaKBBBagBCBmsHBaagnnBaBaB pB ef Per Oeat Class of Aqueous PU BX- Compound Beasent Solution traoted Bydroxy keto ^uinalizarin (with 2.11 95.1 oompounds aniline) 3.80 70.3 4.63 94.8 5,54 95.5 Benzoin 2 to 8 2.2 B-Hydroxybenzophenone 2.6 to 6.3 2.6 8.3 18.6 E-Hjrdroxy-S-methyl- 1.5 to 6.2 0.3 beazophenone 8.47 10.1

•/-Biketones Benzoylaoetone 3.75 3.9 4.«« 7.0 5.47 7.0 8.52 25.0 Dibenzoylffiethane 1.80 0.7 4.20 1.5 4.80 U.O 5.70 36.0 -4?-

Table 2 - Continuad

pE of Per Cent aiass of Aqueous Pu Sx- QompavmA HeagoBt Solution traeted Trifluoroaoetylaeo- E.5 20 tone (witb beazeae} 3.5 36 4.5 57 5.0 86 5.6 93

Oximee and Ni« -st'-HitroBO-y?^- napiithol 2 to 8 <10.7 troso Compounds . ^-NitroBo- ^-aaphthol 2 to 6 <7.6 CHB^drozyaaetopI^aone 2 to 4 <0.1 ctxima 6 13.0 8 2.7 cK -Benzoin Oxim* 2 to 8 <3.7 Xsonitrosoaoetophenon# 2.6 to 5.8 <1.9 3.45 44.2 Isatin-P-oxlmo 1.7 to 6.7 <2.5 Dimethylglyoxime 2 to 8 <4.2 Aoetylaoetone dioxime 1.5 to 8.0 <1.5

MonoSmlBes Aoetylaoetone monoimide 1.8 to 5.3 <1.0 Salioylalaethylimine 4 to 8 <8.4 Salioylalethylimine 4 to 8 <6.0 Salioylal- ^^bydroxy- 4.9 to 6.6 <1.0 ethyl^-iffiine Salioylal phenyliraine 2 to 6 <7.2 Salioylal-^-iiydroiy- 5.68 42.5 pheny^-iffline 5.94 47.3 6116 49.6 6.44 49.2 -43

Table 2 - Continued

of Per Oeat Glass of Aqueous Fu H- OoapouB^ Reageat Solution trapted o-^ydroxyazo 2-Hydroxy-5-methyl- 3.3 to 5.6 a.o impounds azobenzene 1-Be nzeneazo * 2-iiy» 2.95 to 8.70 <1 droxynapiithalens 2-Hltrobettzeneazore- S to 6 <5.4 sorolnol 8 14.0

Hydrozamio Benzobydroxmale 1.86 to 5.20 <21.5 Aeids aoid S.ftt 29.4 6.oe 42.1 6.40 41.9 6.72 48.7 7.61 65.5 8.02 60.0 o-Hydroxbenzohydro- 0.98 to 5.97 <3.6 akle aeid -Furobydroxamlo 3.26 to 6.72 <5.9 aoid 7.92 16.0

Bitblooar1»a.- Mmonium phenyldi- 3.76 to 8.18 <1.0 Biates tlilooarbamats AmmoaltBa morpholine- 1.55 to 5.45 <1.1 dithio sarbamats 8.38 43.8 Potassim pentamethyl- 4.00 7.4 enedltbiooarbamate 4.95 7.1 5.95 7.4 7.07 14.1 8.95 8.8 8.97 7.7

Potassim N,H-4i- 4.15 0.9 ethyldithiooarbamate 5.05 1.8 6.05 16.9 7.07 6.0 8.03 2.2 9.01 7.2 -44-

Table Z - Continued

pH of ?e7 G«it Olass of Aqtieous Pu Sx- Ck»mpotmd Reasent Solution traeted ^Ide Oxlmes Beazamlde oxide S.OO to 8*08 ^.5 Pbenylaoetaiaide ozlme 3.12 11.5 4.30 6.8 5,00 8.3 6.10 21.6 7.00 8.9 7.95 11.6

Misoellaneous Dipbenylthiooarbazoae 4 to 10 <1.0 Bldentate Beageats sym.-Dtpheaylthiourea S to 6 ^4.5 e<-Fieoliaio aoid 2 to 8 ^1.0 Dipheayloarbazide 4.55 to 6.94 <6.7 Salieylalphenylhy- 3.04 to 7.06 <1 drazoae

o-Hydroxyaoetophenone Z 2.1 'semioarbazide 4 5.5 6 14.1 8 15.9 S-Hydroxyquinoline 3.00 2.3 4.04 2.3 5.00 5.8 5.90 19.7 6.86 38.9 8.10 51.3 o-^i&opbenol 1 2.6 6 4.3 9 14.3 Aathranilio aoid 3.9 to 6.1 < 2.3 H-Methylsalioylamide 1.84 to 8.48 <8.8 45

fable 2 • oontlnued

pH of Per Cent Class of Aqueous Pu Ex- Oompouad. Reagent Solution traoted -Azobenzene-/-keto 3*20 to 5*55 ^ 1*0 butyrie aoid N,N* -Diphenylfontt' 1.82 to 5.92 < 1.2 azylbenzene

from the results listed in Table 2 there appears to be definite eTidenoe of oomplexing of Pu(I?) by quinallzarin, the /-diketones, benzohydroxamio aoid, 8-hydroxyqulnollne, and salioylal-/5-hy4roxyphenyl7-imine.

5.1.5 Experiments with quadridentate reagents

It was BTident from the results of the experiments with bidentate reagents that the ooordination number of Pu(I7) is eight. Therefore, it appeared worthwhile to investigate the properties of quadridentate reagents containing two aoidio groups and two coordinating groups per wolecule. If proper re- aotivity and spatial relationships could be found, quadriden tate might be xaore likely to complex Pu{IV) than bidentate rea gents. negative results were obtained with the ethylenediimines of o-hydroxyacetophenone and acetylacetone at pH*s from 2 to 8.

Exploratory experiments v/ith disalicylalethylenediimine, how —46—

ever, gave promising results. The generalized structure for suoii compounds would be represented in the figure below, where M represents a quadrivalent metal.

H H I { C « M^- (OHgL - H • 0 - I -V n ° 0 — 0 - X/ 2

The results of a series of experiments designed to determine the effect of changing the size of ring II are given below.

Table 3 Behavior of Tracer Plutonium (Pi?') with Derivatives of Salicylald^hyde in i^cetate-Buffered Solutions

pH of Aqueous Per Gent Pu Addendum Solution Extracted Hydrazine 2 to 6 <9.4 Me thylenedi imine S to 8 < 3.1 Ithylenediimine 4.59 20.5 5.00 76.6 5.52 93.0 6.01 96.6 Trimethylenediimine 4.38 25.0 4. 90 12.5 5.40 1.3 o-phenylened iimine 4 22.5 6 14.0 a 3.5 47

The results shovm in Table 3 Indloate that chelation is accomplished most easil;f with the ethylenediimine, so the next efforts were directed toward finding derivatives of disalicylalethylenediimine which complex Pu{17) at pH*B below that at which the hydroxide precipitates. The results of several such experiments are siiiamarized in Table 4.

Table 4 Extraction of Plutonium{IV) Tracer by Chloroform and Schiff's Bases of Ethylenediamine

pH of Per Cent Aqueous Pu SJC- Parent Aldehyde Solution tracted

S-Hydroxy-3-methoxybenzaldehyde 2 0.3 4 54.4 6 14.0 8 1.5 2-Hydroxy-5-methylbenzaIdehyde 3.20 11.8 4.25 22.8 4.82 31.3 5.38 57.0 5.68 87.0 6.09 97.4 2,3-Dihydroxy-6-phenylbenzaldehyde 1.12 0 2.85 43.8 4.00 56.8 5.12 59.1 5.55 67.2 5.98 86.0 -48

fable 4 - Continued

pH of Per C«Lt Aqueous Fu £x- Fareat Aldehyde Solution traeted

2-S3rd7ozy-3-ni tro benzaldel^de 3.3S 4.1 3.98 5.2 4.SO 1.7 4.98 2.8 5.39 2.0 5.95 11.8 S'.^ydrozy-S-bromobenzaldeliyde 2.61 28.8 3.30 49.5 3.48 47.2 3.88 26.0 5.05 20.2 5.30 34.2 5.80 39.7 6.30 75.5 g-ardroxy-S-bromo-S-tert.-butyl- 1.98 0.8 benzaldebyde 3.12 36.0 3.22 40.4 3.59 71.0 3.95 39.5 4.53 43.0 5.09 49.0 S.19 54.0 5.50 65.0 5.90 86.0 5.98 92.0 S-Sydroxy-S-ohlorobenzaldebyde 2.33 26.2 3.30 65.5 3.69 65.5 4.20 50.8 4.55 42.8 5.02 47.0 5.31 58.0 6.15 82.0 -49-

Table 4 - Contlaued

pH of Per Cent Aqueous Pu Sx- Parent Aldelurie Solution traoted a-Bytroxy-S-ohlorobenzaldehydo 2.29 25.4 2.88 82.0 3.36 69.5 4.48 57.2 5.45 75.6 i.OO 71.4 8-]^drozy-3 f S-dioliIorobena^ldehjd* 2.48 54.4 3.2ft 83.9 3.92 60.0 4.55 53.8 5.00 63.0 5.65 71.0 E-Hydroxy-3-oliloro-5-t®rt.-butyl- 2.08 59.0 beaaaldehydft 2.75 97.1 3.62 99.5 4.12 99.4 4.61 97.1 5.66 94.0 B-i^irdroxy-&-iaopropyl-5-oliloro- 3.61 46.0 6-^etliyXbeazaIdekyd« 4.70 81.0 5.23 89.0 5.82 89.0

2-Hydroxy-3,5-diaethylbeazald«hyd® 2.72 50.7 3.90 61.8 4.38 45.5 5.00 61.7 5.48 71.4 5.98 8516 2-HFdroxy-S-olilop©-4-t®rt. -butyl- 2.23 58.2 bensaXdehyd® 2.91 83.3 3.85 71.8 4.59 52.4 5.16 75.6 5.75 92.6 -50-

Tabla 4 - Continued

pH of Per Oent Aqueous Pu lJC% Parent Aldehyde Solution traoted

2-Hydroxy-5-tert.-butylbenzal- 2.69 41.0 de&yde 4.52 89.0 5.16 91.4 5.4S 91.C 6.08 84.7 6.98 21.0 B«Hydroxy-5-9hloro-6-methylbeazaX- 8,75 9 deiiyde 4.30 17 4.91 37 5.92 58 E-Hydroxy-5-tert.-amylbenzaldekyde 2.20 1.3 2.62 6.9 5.S2 54.2 4.62 69.3 5.28 80.0 6.14 77.7

S-ii^droxy-4.0-diiiethylbeQzaldehydtt 4.OS 4.5 4.91 31.4 5.47 85.7 6.22 84.7 2-Hy

Table 4 - ContiQued

pH of Per Geat Aciueous Ptt Sx- Parent Aldehyde Solution tracted a.3-Sihydroxy-5 or 6-tert.-butyl- 1.99 9 benzaldehyde* 1.98 83.5 8.28 86.9 2.75 9?,2 £.90 88.5 2.95 99.9 3,95 96.6 3.98 99.6 4.42 99.7 4.64 99.4 4.92 9S.0 5.02 99.1 5.30 96.8 5.60 99.2 5.96 97.4 6.40 91.0

5*1.4 Biaimssloa

^ere are sereral faotore to be oonaidered la tj^eer studies on inner oomplexes of plutonium. The stability of wmy organic oomplexes is quite oritieally dependent on the PH of the medina in whieh the eomplexes are fonaed, There- •The position of the tert.-butjl group is not definitely kn0m* "Rie aldehyde was prepared by the Duff reaction from 8-hydroxy-4-tert>-butylphenol, and the tert.-butyl group in the aldehyde might be in either the S* or 6- positioa. Positive proof of structure depended on the oxidation of the tert,-butyl radical to a earboxyl group, and tertiary alkyl radieals are particularly resistant to oxidation. However, the sore likely structure might he the S-substltuted compouM, as Judged froM the reactivity of other 5-substltuted coispounds with Pu(IT). -52

fore. In order that no reagent be orsrlooked in its possi bility of ooMplexing plutoaium, a rather wide pH range was iiivuistifeuiyu, iiie ouiTer syeteius were limited to acetates, beoause the anions present in other buffer systems (citrate, tartrate, eto.) form very strong complex ions with Pu(I?), and might lead to erroneous oonolusions about the beharior of Pa(r?) with some organic reagents. fraoer soale extractions can lead to results which a re of great Talue, even though some of the data may not be con- olusiTe on a quantitative basis. If there is high extract- ability into a solrent which vsfill not extraot plutoniiaa in the absence of the organic reagent, the results can be taken as presumptive proof of inner complex formation. If the ex- tractability is low (less than Z per cent) it is not li&ely that oo^plax formation takes place. Low but still laea&ur- able extractabilities (less than 5 or lOj^) are almost ambi guous in their meanings. Many of the compounds which gave promising results on the tracer soale were later studied more thoroughly on the micro soale using visible amounts of plutonium. fhe high extractions obtained with quinalizarin and aniline are difficult to interpret, beoause aniline alone ef fected about 85 per cent extraction of tracer Pu(I7). This may be due to selective wtting of the hydroxide which might be precipitated at the pH produced by the solubility of ani -53-

line in water. Quinaiizarin -with ether, tert.-amyl alcohol, and aitromethaae effected no extraction of Pu(IV) tracer. The estractlon of tracer FuCIY) hy tannic acid smd aniline (53)

haa been reported, but there has been no elucidation of the

cheiaistry involTed. Later, in connection with ezperiiseats designed for other purposes, it was found that micro amounts of I»u(IV) formed a reddish-purple complex at pH » 6 with quln- alizarin, and that the compound was soluble in and extractable from water by eyclohexanone. IHie /-diketones had obTious possibilities as chelating compounds, since acetylacetone long has been known to complex many metals, including Th(IV) (54), Zr(lV) (56), U(IT) (56) and Ce(r7) (56). Ck>nsiderable work has been done by other laboratories on Pu(IV)-aoetylaoetonate (57). In our work we did not stress the study of ^-diketones because as a rule their coaplexing activity is quite general for many eleiaents. However, it was found later by OalTin and co-workers, that certain trifluoromethyljf-diketones exhibit some specificity for forming stable Pu{IY) complexes (58, 59, 60, 61, 62). Hone of the nitroso compounds or oximes possess coiapleoc- ing activity for PudT). Gowaa and Goldsmith (63) also ob served negative results with #^-.nitr080-^->naphth0l, dimethyl-

glyoxime, and salicylaldoxime on the tracer scale. The only

nitroso compound whose complexing activity with Pu(IY) is im portant is cupferron (64). 54

Of all of the other mlsoallaaeous hidentate reagents listed in Table 2, only benzohydroxamio aoid and 8-hydroxy- q.uinoline showed appreoiable oomplezing aotirity vdth Pu{IV). Barly experiments (65, 66) had indioated that the 8-hydroxy- quinolates of Oe(III) and Ce{IYj when preoipitated at a pH of 4 carried plutonim. Considerably later, Fattonfts analysis of Pu(IV}->S*hydroxyg.uinolate on the mioro soale gare eonolusive piroof that the ooordination number of Pu(r7) is eight (67)« Many of the bidentate reagents listed in fable 2 are re markably aotive in forming inner ot^plexes with many metals, especially those in the so-oalled transition groups. That so few of them reaot with Fu(I7) may oonstitute evidence that Plutonium does not resemble the transition elements in whioh d orbitals are being filled. The results obtained with quadridentate reagents are interesting in that they show the structural relationships necessary for the formation of complexes with Pu(IV). Tracer soale experiments with di-aeetylacetone ethylenediimine gave negatiTe results, and attempts by others (68) to prepare the U{IV) and Pu{IT) complexes, by the method used by Combes (69) for the preparation of the Cu(II) eomplex, were onsueeessful. •JHae results swmarized in Table 3 ishow quite well the effect of chelate ring size on the stability of the Pu(IV) -55-

Qomploxes with disalioylaX diimines. Copper and niokel deri- TativeB of this type hare been studied quite extensively by Ffeiffer (70)» and he has described oomplexes in whioh there €ure as many as ten methylene groups between the two imino ni trogen atoms, Pfeiffer (71) prepared oomplexes of di- salioylaldiiainee, and the U(IV) complex of disalioylalethyl- enediimine is stable with a melting point above 300*^0 (72). Aetually for oompounde of this type, little Quantita tive work has been done on the correlation of atruoture with stability of oomplexes with disalioylaldiimines. The reoent studies of Nuffield and Calvin (73) on the exchange reactions of copper chelated compounds mark a step in the right direc tion. The ethyleiMdiiiaines exhibited the greatest tendency for o

5.2 Extended Stiidlee with Diaal^

§.8«i Oonditioas for preparing the Ptt{IT) oomplex oa the miero aoale

In experimente on the tracer scale, disal proTed to he the most promising of the reagents exeuained for complexing activity with Pu(I7} in the pH range Z to 6.4. Because tra cer scale experiments can sometimes lead to questionable re sults it was necessary to check the validity of the results with aicrograa and milligram amounts of plutonium. Several experiments were performed to determine the op timum conditions for preparing ths eoraplex with micro aiaotints of PuClYj. In all of the experiments, the Pu(I7) was in 1 ml of acetate-buffered solution at pH 4. The solid reagent was added (usually twice the theoretical amoxmt needed) and the slurry agitated while complex formation took place. During the course of the reaction, the solid phase (the suspes^ed organic reagent) changed from the yellow color of the reagent to the purple black color of the Pu(IV) complex. Chloroform extractions, with 1- to 2-ml portions of sol vent, were done in the micro extraction apparatus described in Section 2,E. The results of several experiments are summarized la fable S. •^e word '•disal^ will be used in the rest of this thesis as an abbreviated notation for di-{a,3-dihydroxy«5 or 6-»tert.- butylben2al)-ethylenediimine« ——— 67-

Tabltt 5 fhm llxtraotion of Plutoniuad?} on the IHoro SoaXe by Chlorofora and Msal from Aoetate Solutions at pH 4

Reaotlon Coaditloae Per Ceat Pu ^ouat of Q Sztraeted Pu(I7} Temp. 0 Time, Brs. iate Ohlorofora

0.950 Bg - 30 S6 ^.4 X.019 25 - 30 48 96.1 0.356 eS - 70 2.5 88.8 0.356 95 2.5 97.5 0.35ft 95 1.0 96.5

The relative slowness of the reaotion is doubtless due to the faot that the reaotion takes plaee in a t^^o-phase sjstesi. The reagent is quite insoluble in water.

5.8.2 The eoiiLposition of the Pu{lT)-disal &omplex

Bisal is a quadridentate reagent with two aoidie and two ooordinating groups, and it was assumed from the re sults of traoer experiments that two moleeules of reagent were involved in oomplexing one atom of Pu{IT). To estafe- jlish this with oertainty, the weight percentage of plutonim in the disal oomplex was determined. To an acetate-buffered solution containing 3.79 mg of Pu(I7) at pH 4 was added an amount of disal less than that required for reacting with all of the Pu(IV) present. The suspension was agitated for 12 hours at 80 to 90®C for coni- 58 pletioa of the reaotlon* The organic o<®plex was extracted with chloroform and the organic layer washed with water. The chloroform solution of the complex was evaporated and the residue dried to constant weight at 100®G. The com plex, almost hlacJc in color, was weighed, destroyed by di gestion with HSOg, and the reeultiag solution assayed for Plutonium. The data are sumiuarized in Table $. The formula ^ (where R « aisal radical) is close enough to PuSg to indicate that two molecules are inTolred with one atom of Pa{17), No molecular wei^t determinations were attempted.

Table 6 (imposition of the Plutonium(17)- Disal Complex

?lelght of plutoniuii taicefi a.790 ng Wei^t of dried complex 11.502 mg Weight of plutoaii£B in ec»aplex, by radioactiTe assay S.569 Bg formula of complex, calculated from the above data ^1.98 where R* is

H CHo ^ y\ ' ' ' ' /V {GHS)3C ->0(0^)3

L 0 - • 0 -\/ ( I 0 0 H a "'Aeawalm the iert.-butyl group to be la the &-positloa. -59-

use of Tar logs solvents for dlsal extraotlone

In aost of the traoer soale extractions ohiorofora was used as the solvent. If there should erer be a need for a solvent more adaptable than chloroform for continuous coun ter current extractions» the behavior of various solvents toward the Pu(X7)-di8al complex would have to be kaoi^. The data sufflBiarized in Table 7 v/ere obtained by using the usual tracer extraction techniques, using 20 ml of aque ous solution at pH 5.6 and 20 ml of solvent in separatorjr funnels.

Table 7 Extraction of Tracer FlutoniuBi{I7) by Disal and Yarious Solvents from Acetate ^lutions at pB 5.6

Class of Per Cent Pu 0om|H>uB4 Solvent Sxtracted H^arlQB

Hydrocarbons Pet. ither 95.0 Preferential (35-5§®C> wetting of aiisal Bensen* 98.9 fblueM 96.0 99.4 Chlorinated Chloroform 99.5 Hlydrocarbons Carbon tetrachloride 50.6 Trichlorethylene 86.9 Alcohols Benzyl alcohol 96.4 Cyclohexanol 98.1 Considerable 2ntersolu bility. -60

Table 7 - Continued

Class of Per Cent Pu Compound Solvent Extracted Eemarks

Ethers Si-isopropyl ether 99.1 Diethyl ether 98.9 leters n-Butyl acetate 99.7 Ketones Methyl isobutyl ketone 99.0 Cyclohexanone 99.3

Tbe results indicate that several solvents are effeotive for disal extraotioas of plutonlum.

5.S.4 Behavior of the Pu(IV)-diaal oomplex toward dilute aoida

Several experiments were conducted to determine the ni tric aoid oonoentration necessary to extract tracer Pu{IT) from ohloroform and methyl isobutyl ketone solutions of the disal ooBiplex. The results of these experiments are summar ized in Table 8. In all oases, the disal extractions were done under pH conditions which lead to virtually complete extraction of Pu(I7). Copper(II) carrier was used to obtain a qualitative ideu of the relative stabilities of the Ou{II)- and pu{IV)-disal complexes tov?ard nitric acid. Table 6 Stability of Tracer Flutoalam(X7)-DisaI Toward Dilute Nitrio A41d

Per Oent Molar Per Oent Total Pu EK- ity of Pu lxt*d. Per Oent SolTeat Carrier PH traoted mo^ by flHOg Tield Remarks

Chloro- Ou*"*- 3.91 99.7 0.05 83.7 82.6 Cu*'*' oomplex stable fora S.9g 99.0 0.2 99.2 97.9 19 ft ft 3.96 96.7 0.4 99.6 98.6 Ou** complex de stroyed 3.95 98.7 1.0 99.5 99.2 Reagent attacked Methyl Hone 5.6 99.6 0.1 94.8 94.4 The reagent was par isobtttyl 5.6 99.8 0.2 95.2 95.0 tially attacked by kietone 5.6 99.5 0.4 97.8 97.4 all of these concen 5.6 99.4 0.6 93.6 96.0 trations of acid. The 5.6 99.4 0.8 96.6 96.9 solubility of BHfOg in 5.6 99.4 1.0 98.6 98.1 the ketone is oonsiderable -62-

la ordftr to determine the effeotlveness of oxallo aoid in extraotiiig Pu(I7} from the chloroform solution of the di-> sal complex, tracer Pu(IV) was extracted with disal and chlo- rofozm from, aoetate-huffered solutions at pH 4.6. Then the chlorofoim solution of the complex was agitated with an equal •ol^e of dilute oxalic aold solution and the two final frac tions were then analyzed for plutonium. ^'ith 0.04 M oxalic acid, 53.5 to 58i.6 per cent of the plutonium was extracted free the ohlorofoxm solution, and with 0.10 M oxalic acid the extractions ranged from 58.2 to 71.4 per cent.

S.a.S Interfering substances

Sulfate, oxalate, and ferron {7-iodo-e-hydroxyquinoline- 5-sulfonic aoid) were found to prevent complexing of Pu(IV) by disal. fhe results of several tracer scale extractions with equal volumes of organic and aqueous phases are suianiarized in Table 9. -63-

Tablo 9 Interferenoes in the EztractioB of Plutoniiaa(IV) by Diaal and Chlorofom

pH of Aqueotafi Per Cent Pu sxtraoted Suhstanees Added Solution by Ghloroforft

Hone 5.2 99.0 0.1 M Ma2S04 5.2 91.6 O.S M Sa:^4 5.2 81.2 0.3 M n&^A 5.2 74.5 0.4 M Sai^l 5.2 70.1 0.6 H 5.2 68.1 0.8 M BasS04 5.2 70.9 1.0 M Ra2S04 5.2 70.4 0.63 M iraES04 1.90 4.9 2.56 4.8 3.17 23.2 3.81 61.0 5.22 87.8 0.4 M 090A' 2.68 to 5.00 5.2 ?erron (0.5 log/ml) 2.02 17.8 2.51 53.6 3.01 49.0 3.52 51.9 4.00 62.5 4.51 53.6 5.00 42.9

^e data indioate that at pH*8 at which disal extraotions are eomplete from aoetate solutions, the Pu(I7} oomplezee with SO^*, OgO^" and ferron are strong enough to limit the ocaaplex- ing of Pu(IY) by disal. "-64-

5.2»fi Dlseusglon

Coaposltioa of the Ptt(iy)*dlsal eompXei:. The ehemioal evldenoe obtained by analysis of the Pu(rf)-disal oomple* indioates with oonsiderable eertainty that the coordination number of ?u{nr) is ei^t. In this respest the behavior of Wnilf) is analogous to that of U(IV), which forms complexes with disal (72)• 7h(Z7} and Ce(IT) also form chelate com pounds ccaaipatible with a coordination number of eight for the metal (54, 56, 74, 75). Very little is icnown about the stereocheaaistry of com- potmdfi of elements with a coordination number of eight, even though many such compounds are known* The isomer tables of Marchi, Fernelius and McReynolds for coordination number eight C7€) indicate that complete configurational studies of coordination number eight by chemical methods would be a formidable task. Beard and Nordsieck (77) hare shown by z- ray studies that the ^olybdooctacyanide ion is dodecahedral, and recently Marehi and McReynolds (78) have resolved po- tassiua tetra-oxalato-uraniuiii-(IY) into optical isomers, and postulate that the oxalato-uranium ion is either antipris* matie or dodecahedral. The studies of chelate compounds of Pu(r7) were designed to concentrate on chemical properties, so there were no at tempts to study stereochexaical configurations, Eowever, it 65 might be interesting to speoulate briefly about the stereoQheaiioal properties of Pu(I7) oomplexes with re lation to the position of plutonim in the periodio table* Present opinion tends toward the view that pltiton- im is a meeiber of an *aotinide" series in v^ioh the 5f aubshell is being filled. On the basis of this Tiew, the configuration of Pu(IT) might be 5 f*, with the 6d, 78, and 7p orbitals raeant. Thus, to form a complex ©OTI- pound with coordination number eight, Pu{I¥) must aooept eight pairs of electrons. These are available in two disal radicals. There are no known oases in which f or bitals are involved in bond formation (26), so it is not lilcely that any of the eleetrons donated by the reagent would be aeoOAimodated in the 5 f orbitals. However, the 6d, 78, and 7p orbitals are available, and mi^t receive the electrons to form stable bonds. According to Sim- ball iZ6)f in his analysis of coordination number eight, bonds could be formed to give the antiprisaatie 4K)nfiguration, but bonds would be much more stable. These would lead to antiprismatic or dodeeahedral con figurations. These predictions were substantiated by ehemical studies on tetraoxalati-o-uraniuia-dV) (78). In order to obtain cubic structures, d^fsp® or bonds would be necessary (26). 66

Magnetio measurements would be of great value la determiniiag whether 5f orbitals are involved In the formation of Pu{X7} complexes. Ghemioal studies designed to determine the configuration of Gomplexes might be quite difficult. Chgaical properties of the Pu{IY)-disal coBiplex. The reason for the slow reaotion of micro amounts of to form the disal complex is probably that the reaction raedium is heterogeneous» due to the low solu* bility of the reagent in water. The reaction rate could undoubtedly be increased by the uee of mixed sol vents, such as dioxane-water mixtures, but this would lead to decreased efficiency in the extraction by chloro form. The data suoimarized in Table 7 show that several solvents are available for dlsal extractions if solvents with particular physical properties should be needed. From the behavior of Pu{17) and Cu(Ii: ) disal com plexes with dilute nitric acid, one might surmise that even though several elements were complexed by disal a fractionation still raight be possible by Judicious choice of the acid concentration used to destroy the complex. In fact there are considerable differences in the sta bilities of metal complexes of coaipounds of this type. Pfeiffer and co-workers (79) have studied the stability of complexes of disalicylalethylenediimine with various 67-

^lementa by xQeans of displaeeaent reactions and haT« foimd the relatlTe strengths of complexes to be In the order Ctt>Nl >V, Fe >2ai, H >Mg. Thus, It Is pos sible that the specificity of disal for plutonium might be increased by proper use of an acid re-extrao- tion eyele. The low re-extractions of Pu(IV) by oxalic acid were surprising in view of the fact that oxalate inter feres so strongly in the complexing of Pu(IT) by disal. See Table 9. It is interesting to note that oxalic acid in concentrations as low as 0.05 M will remove Pu(I7) from the cupferron conplex (80), The interferences of sulfate, oxalate and ferron with disal extractions are ascribed to the fact that these substances form very strong water-soluble com plexes with Pu{IY). The Pu(IV)-ferron complex is so stable tdiat ferron prevents the adsorption of Pu(I7) on Amber lite IH-1 resin (81). Attempts to destroy the Pu(IV)- ferron compound by addition of ?e*^^ ion to form the more stable ye(III)-ferron did not lead to any success in im proving the yields in disal extractions tvom. solutions containing ferron.

5.3 The pu(IV)-Ferron Ocaiplox

5.5.1 Introduction

Ferron (7«iodo-8-hydroxyquinoline-5-sulfonio acid) -•68—

foxms aa extremely stable complex witli PutIV) in tlie pH range 4 to 6. Tlie formula of the oomplex is pro bably

SO^H I

Fu 4

Ihea ferron is present ia a solution of Pu{17) at the proper pH, oomplex formation preTents adsorption of Pu(Hr) by Amberlite IR-l resin. This makes possible the separation, ia the adsorption process, of plutonium from elements not oomplexed by ferron (SI). The work described in this section was designed to learn some of the chemistry of the Pu(I?)-ferron complex, especially that of analytical importance. In 193E, Toe (62) reported a method for the colorl- metric determination of ferric iron with ferron« Tery few cations (only copper, cobalt, nickel, ehromiuai and aluEiintim) were reported to interfere. Later, Toe and ^11 (83) and Swank and Mellon (84) published additional data on the ferron laethod for iron. They stated that Tery few ions other than Fe(III) form colored oomplexes with ferron. Therefore when Ayres (81) found that Pu(IT) formed -69- a stable ooiaplex with ferron, it was of interest to d»- termine whether the oouplex was colored, and whether it might hare any peculiar properties which night be of use, in the analytical ch^istry of plutoniuB, In the experimental work described below all of the spectrophotometric laeasureEients were made with the Beck* man Model HV Speotrophotoraeter.

5.S.2 Stperimental

13ae absorption curves for ferron* In laost speotrophotometric work the obtical behavior of the solution stu died is compared with that of a "blank" solution. la stu dies with ferron, for ezaiuple, the sastple would contain ferron, a buffer, end the cation under consideration, while the '•blank*' would contain only ferron and buffer. "Hie eolor Gharacteristies of ferron itself ohange with pK, so spectrophotometric curves for the reagent are necessary to determine how much care must be exercised in preparing th» «blank solutions'*. ^e absorption curves for ferron are shown in Figure S. All of the measurements were made with 1 em cells filled with a buffered solution 0.000228 M in ferron against a blaoJc containing the buffer* It is obvious that in the re gion below 500 m.f/es-Te must be exercised in preparation of the blank solution. -70-

3000

2000

a 1000

300 405 500 ^0 Millimicroas

Pig. 3. Absorption Curves for Ferron at Different pH's A: pH 8.31; B; pH 3.06; C: pH :^.36 D; pH 5.49; E: pH 7.10; F: pH 7.98 -71

Absorptiog ourves for the Ptt(I7}-ferron eoapXex. la Figure 4 are shown absorption ourves for the Pu{17)- ferron complex at pH*s 3.61 to 4.88. la all eases, ferron was added in ezeess of that required to eomplez the Pu{IT}, and the blank contained the same buffer and aa amouat of ferron equal to the ezoees used in the pre« paration of the complex. Two of the curves were obtained trwa solutioas of plutoniim purified by peroxide preeipitations. The curve at pB 3.61 was obtained from a solution containing 20.3 (Wg Pu{I7)/ml. The solution had a very slight greea co lor. At pH 4.68, the ferron solution containing 4.63//g Pu(IV)/ml had a faint greenish-yellow color. The third curve, for a solution of Ptt(IT)-ferroa at pH 4.88 containing 17.0^g Pu{IV)/ial, was prepared from Plutonium by fluoride precipitation. ^e absorption ourves for the ge{III)-ferron eoaplex. Jn yigure 5 are shown the absorption curves for the Fe{III)- ferron complex, as calculated from the data of Mellon aad Swank (84). They measured traELsaittaneies against a blank containing only the buffer and no ferron. Therefore, the ourves below 500 m^are not particularly useful, because ferron itself has aiariced absorption peaks in the region 430 to 440 However, the peaks in the region 600 to 620 m/t/are certainly due only to the FeClII)-ferTOn com plex. 6000

® 4000

eq 2000

TO ^0 Millimicrons Fig. 4. Absorption Curves of the Pu(IT")-ferron Complex at Different pH*s A: pH 4.68 B: pH .3.61 C; pH 4.88 Molar Extinction Coefficient

cn CJI

cr 0) o •aCt- O » CJp <3I •1 03 < I aM o

I 9 *i 9o 0 1 9H 'I SpeQtit)phot

pH 2.77

pH 2.38

0.2 pH 1.96

M PH 1.69 o :sM I PH 1.50 0.1

pH 1.02

400 500 600 700 800 Millimicrons Figure 6. Absorption Curves of Pu(IV)-Fe(III) Solutions Containing Ferron -7e-

Tatjle 10 Spectropbotometrio istl/iiation of Iron la Plutoniuai with Ferron as Reagent» Using the 600-620 n// Band

pH of of ?e(III)- Log 1q/1 from Solution ferron ?ig. ft g Fe/ml l.OS •» 0.038 l.SO * 0.106 1.99 1900 0.1S4 4.9 1.96 2950 0,183 4.0 2.36 3000 0.S25 4.S 3300 0.250 4.2

^liiaearity between and pH no longer holds. B

5.3.3 Diseimsioa

There are iadloations that the ferron method oaa be ddapted for the estimation of iron in plutoniua, because Pu{IV)-ferron has no siarked absorption in the 600 to 620 m// region of the oharaeteristio peak for FeCIII)-ferron. Hie absorption ourTes for Pu{IY)-f0rron in Figure 4 iadieate that some iron might be present in the pero xide-purified plutonium, because of the slight inflee- tions around 590 m/i/, The corresponding inflection in the cuTTe obtained froia fluoride-purified plutoaium is not so marked, Indies ting less possibility of iaron con- taraination. The Inflections at about 450 are not 80 -7?- sigaifleant, beoauae ferron Itself bas an adeorptlon peak in this speetral region, end the infleotions could have beea due to slight errors in blank eorreotions. -78

n •THE CHELATE CC&!I>OUNDS OP PLUTOHBMdll)

6.1 Introduotioa

Siploratory attmpts to produe® traeer Pu{III) did not proTe partioularly suooeesful, •speolally at high enough for the fomatlon of stable orgaaio oomplexes. The anions present in the oomon buffer systems eoaplex PudT) quite strongly^ and oonsequently shift the potentie^. of the Fu(ZXI}*Pu(X7} eouple. Therefore, on the traoer scale, even though proper reduoing conditions have been used for produc tion of Fu(III), there is always xmoertainty about the iden tity of the oxidation state of plutonium in the buffer sys tem. for these reasons, it was necessary to conduct all of the exploratory experiments with micro amoxmts of Pu(III), 8Q that oxidation states oould be identified ndth more cer tainty by spectrophotcHaetrio methods.

6.2 Experimental fe.a.l General procedure

jfi the exploratory experiments, the amounts of plutonium used Taried from 106 to 500 y6/g. In the Tolumes of solutions -79- used, these quantities of plutoaima usually were suffi cient to make possible the identifioation of the oxida tion state by speotrophotometrio analysis {8S). For identification of Ptt(III) in solution, the absorption bands at 560 and 600 a^were used. When reduotioas were oomplete, the oharaoteristie 480 m^baod of Pu{17) was ab- s^t. All speotrophotometrio measurements were made idth the Beekman £fiE>del W speetrophotometer. After speotrophoto metrio measurements had shoim that the reduotion prooedure was effeotiTe it was assumed in most of the suoeeeding ex periments that Pu(IXI) was the oxidation state present in solution. Later, however, in the determination of the eom- position of Pu{Z7)-^-naphthohydroxamate, it was found that in aoetate solutions, oxidation to PudT) aay take place to a certain extent. The usual reduotion procedure was to remoTe most of the nitrate present in the aliquot of stock solution by sev- ex^l evaporations with HGl. The plutonium solution then was made 1.0 to 1.5 H in hydroxylammonium ion and 1.0 to 1.5 M la HOI, The tiae for reduction was alwftys at least an hour. In general, the volume of solution was kept below 1.5 ml. Tn the moBe concentrated solutions, the color change from brownish green to blue was evident as reduotion to Ptt(IXX) occurred. Ihe organio reagent, either as the solid or in ethanol solution, and in five to ten-fold excess, was added to t^e Pu{2II) solution. Ilie theoretical aaounts of reagent were -80-

basad on the assuaption that th« doordiaatioa nmbet of I'uClII) is six. After addition, of the reagent, the pH of the solution or suspension was raised gradually by addi tion of HH^OH and ammoniuci aoetate until the aoetate eon- eeatration was about 1 M. At least two hours were allowed for reaction to take plaoe.

Chloroform extractions were made in the mioro-extrao- tion apparatus with the Tolume of solTent equal to that of the aqueous phase* The fraotions were analyzed for plutonium by eraporating aliquots on platinum foils and oheok- ing their alpha aotivities.

fhe behavior of plutoniucidll) with hydroxamio aolds

It is generally assumed that the tauto^ers (X and XI) of the monohydroxeLffl$Q aoids are eapable of existenoe, but that the oxime strueture (IX} is neoessary for chelate ring formation <86)*

0 H OH II I 1 H - S -H - OH R- O^H—OB I II

1!^e hydroxamio aeids are quite aotive in fonaing ohelate oompounds, and therefore it seemed worthwhile to examine their astivity with PuClII). ^he results of several exploratory experiments are sum- -81 marlaed ia Table 11. la all cases, the experimental prooedure used was that de&oribed in the preoeding seotioa.

Table 11 The BehaTior of Micro Amounts of l^lutonlum(XZI) with Tarious Bydroxamie Acids

pH of Per Gent Pu Extracted Eeagi^t Solution bj Chloroform

Benzohydroxamio Acid 5 43.S « 92.5 m-*llltrobeazoh;^rozamie 6 74.6 Aoid o-BydroxybeBZohydroxamie 6 33.0 A© id -Naphthohydroxaaio Aoid 2 2.8 4 94.3 6 97.4 <^-Furohydroxamlc Add 6 1.2 Phenylaoetohydroxamio Aoid 5 47.2 f 79.5 Acetohydroxamic Aoid 5 1.2 a-Propaaohydroxamio Aoid 6 5.0 a-Yaleroh;?^oxamie Aeid 4 96.4 6 98.6

The data ia fable 11 show definite eTideaee of emplex formation by Pu(XXI} with several of the hydroxamio acids. Ia all oases in which there was extraction of plutoaiom by chloroform, the color of the Pa(III) complex was reddish- orange. The corresponding hydroxaaates of Fe{III) hare a laore aearly purple aolor. In tracer scale extraction experiments with Pu{IV) (see -8£-

^able 2), bensohydroxmic acid showed definite evideaoe of completing Pu(IV). However, it was possible that Bome of the Pu{IV) might have been reduoed to Pu(III) by the hy- droxastie acid, vs'hieh ie really an aoyl derivative of hy- droxylaiLine.

It vas found with i&illigram aBK>unt8 of plutonium In the tetrapositive state that "^^-naphthohydroxaado aoid aM chloroform effected 39.0 per oent extraction of Pu(IV) from a solution of pH 6. Similar experiments «ith Fu{IXI) showed that naphthohydroxamio acid and chloroform effected 97.4 per cent extraction from a solution at the same pH, There fore it was of interest to determine the oxidation state of the Plutonium in the hydroxamates listed in Table 11. The oiamposition of the ^-naphthohydroxamate vias deter mined in a way similar to that used for determining the com position of the Fu(IV)-diBel coiaplex {Section 5.2,2). The <=^-aaphthohydroxaycaate was chosen for the gravimetric analy tical work because of the relatively high molecular weight of the parent acid. Two determinations of the composition were made. In the first experiment, Pu(IIl) was prepared by re duction in hydi'oxylajamoniuffi chloride and the identity of the oxidation state checked spectrophotoaetrioally. This solution containing 2.972 ag of Pu{III) vi&a buffered ia asamonium acetate to pH 4^ and as the pH was raised the eolor of the solution changed from blue to pink. To the -83-

3 ml of solution were added 5.433 mg of X-naphthohydroxasiio acid. A brown color developed almost imir.edlately, and the suspension was agitated with helium for 7 hours for comple tion of the reaction. Extraction was made vdth 3 ml of ehlorofom. The coiaplex dissolved readily in the chloro form to fom an orange solution, leaving the aqueous layer colorless. The chloroform solution of the complex was care fully evaporated in a weighed micro platinum, dish and dried to constant weight. Dhe plutonium content was determined by destroying the organic matter with nitric acid and assay ing the solution for plutonium by alpha counting. In the second experiment, no acetate buffer was used, ffiie solution, containing 2,535 mg of Pu(III), was prepared by reduction in hydroxylaxamoniua ehloride. fo 1 ml of this solution were added 3.648 mg of cX-naphthohydroxamio acid. As the pH %'as raised to 6 by the addition of HH^OH the sus pension acquired alaost iimediately the characteristic brown color of the hydroxamate. The suspension was agitated with a helium stream for 7 hours for completion of the reaction. Extraction was made with 2 ml of chlorofom. The chloroform solution of the complex v'as evaporated and analyzed ia the same way as that described in the preceding paragraph. The results of the two determinations are sho\'m in Table 12, -84

Table IS Ck^mpositiea of Plutoaiua(III)-<7(-NaplitlioiLydroxaiaate

Conditions for the Preparation of the gtoaplex X M Aeetate Ho Aoetate, Solution, pH 4 pH 6

Weight of Plntonium Used S.972 ng 2.535 ag Weight of Organio Heagent 5.433 m 3.648 Mg Weight of Complex Fozmed e.SSO iog 4.9^ mg Weight of Plutonium in Ck)si.plez 2.199 ag 1.432 mg Oomposltlon of Oomplex* PuRg,^

•where H - ^/-naphthohydroxaiaato radioal

V_V

fhe data indioate that Pu(III} is not very stable la aoetate eolations. Sren though in the first experiirient all of the plutoalum wae originally in the tripositive state, addition of aoetate ion may hare caused some oxidation of Pu(III) to PuClT) because of the st-hilizatlon of Pu{IT) Isy formation of the complex acetate ion. The change of Pu{III) to Pu(IY) In aoetate solutions probably becomes more appre ciable with longer tiraie for reaction. In the first experi ment awiiaarized in Table 12, 7 hours were allowed for re action, and in the exploratory experiments suainarized in Table 11 the re&ction period ^as from 2 to 4 hours. Thus, even though possible that acme oxidation to Pu(lV) 85-

aaj nave occurred, it still appears that the positive re sults sumraarizel in Table 11 indioste coinplexins of ?u(III) fey aertain hydroiaiaio acids.

Another faotor had to be oonsiderad ia the reaotioas of hydroxaaie acids. In all of the experiments with Pu(III) and hydroxaaie acids, hydroxylaBiiaonivmi chloride was used as the reduotant. There was the possibility that at the rela tively high pH»8 used for preparing the ooiEplexes, some free hydroxylamine might have reaoted with one of the tautomers of the hydroxamie acid to form the corresponding M,N*-dihy- droxyamidine according to the following reaction

a — 0 = 0 + HgN-OH a - C = HOH HgO ) H- H H-H I 1 0 0 H H

Ixperiaents ^•s?ith benzohydroxamie acid and hydroxylammonium chloride i^ith oonditions identical with those used for III)-complex formtion showed that the hydroiamic acid was unchanged. After 2 hours in the reaction medium, the melting point of the benzohydroxamio acid was unchanged {i29®C). ll,H*-Dihydroxyhenzamidine melts at 115®0. (H. Ley, Ber.. S1E7 {1898).

6«a.5 The behavior of plutohiua(III) with amide oximes

The amide oximes constitute an interesting group of ecxBLpouBds which apparently exist in two tautomerio forms ••86'

H — 0 — KHg R— 0 = NH N-H 0 H la. wiiiek they reamble the hydroxaa-io aoide:

R — 0 —OH R— C =-0 II H N-H 0 0 H H

It was tliouglit that the amide ozimes might also reserable the hydroxamio aoids in their reaotions with Ptt(III). The experimental proeedure for examining the activity of amide oximes ti^as the same as that described above for the hydroxamio aoids. The results of several exploratory experimsntB are smaraarized la Table 13. -87-

pH of Per Oent Pu Sx- Reagent Solution traoted by Ohlort>fora

Benzamide Oxime 6 0.3 j^-foluamide Oxime 4.1 0.9 5.8 13.8 (<-Haphthafflide Oxime 5 1.6 6 10.9 Fhenylaoatamide Oxime 3.0 1.8 4.1 0.2 5.0 98.5 6.0 77.9 n-Taleramide Oxime 4 3.2 5 46.5 6 9£.0

loth phenylaoetamide oxime and a-valeramide oxime eomplez Pu{ZII) quite oompletely to form oompounds doep purple in. oolor. IQL ooaaeetion ^ith some other studiee it was fotmd that phenyl-aoetamide oxime al&o reacted with Pu(IT)* V/hen 826 of peroxide-reduoed Fu(IV) (v?ith no Pu{III) or Pu(?I) present, as shown by spectrophotometric antilysis) were allo«^ to reaot with phenylaoataadde oxime at pH 5 to $, one ohloroform extraction removed 89.2 per eent of the plutonim. The pudirj-phenylaeetaciide oxime is brownish-orange in color.

6.3.4 The behavior of piutonium(III) with misoellaneous bidentate reagents -88-

Tke analysis of Pu(IIl)-*<-.naphtiiohydroxamat® g&7e d9fi- aite eTid«iaoe that the ooordlaatioa aumber of Pu{III) is six, Tliarefore, the type of reagent most likely to form chelate aompouads of Pu(III) is the bidentate reagent eontaining one acidic and one coordinating group, l&e results obtained in cursory experiments with Pu{IIl) and several aiiscellanecus bidentate reagents are summarized la Table 14. The usual experimental procedure was used. Most of the reagents examined exhibit rather general ooaplex- ing action with many metals.

Table 14 The Behavior of Micro Amounts of Plutonium(III) with Terlous MiscellaneouB Bidentate Reagents

pH of Per Cent Pu ®x- Eeageat Solution tracted by Ghloroform

38licylaldoxime 4 0,67 6 0,S5 o<-|litroso-^-naphthol 5 9.8 Dimethylglyoxime 6 1,2 BiphenyiLthiocarbazoae i 3.2 Is a tin-f-oxime 6 5.5 c<-]^nzoia Oxime 6 3.3 fhenylglyoxaldioxiarie e 1.4 Oiph^ylthioearbazlde 5 6,e a-Diphenylthiourea e 6.5 S^Ofdroxyquinollne 5 10,6 6 51,6 AiBmonium Morpholiae- Jithlocarbam&te 6 0.5 Potasalm peatamethyleae- dithiocarbamate 1.8 Potassiiaa N,N-diethyldithio earbaaiate e 0.1 -89

It oan be sees that S-hydroxyquinoline is the onlj reagent listed in Table 14 showing evidence of oomplexing Pu(III).

6.3 Disoussion

^e analysis of Pu{III)-

and ^87), Werner (37) postulated that the structure of the ccaaplexes of hydroiamio aeids Is -90-

SDweTar, it kas been found tliat dorivatives of th® genaral formulas HCOHHOR* or RG{OH) ® NOR* do not give eolor reaotions with f«rria ohlorido, while oompounds of th® strue- tuirea

Q ^R» ^0R» .01 E — e — H , E — C-: ,aadR — C ÔH ^NOH do, so it is believed that the 5N0H grouping is necessary for the fomation of chelate compounds (86)• If this be true* th® hydrozamio acids and amide oximes are the only com pounds with »NOH groupinga which show chelating activity with Plutonium. Phenyleceto- and n-valerohydroxasiic acids are quite sol uble in water, as are the phenylecet- and B.-valeraffiide oxiiaes. fhese solubility properties malce these compounds much xmrm useful as reagents for plutonium. Salicylaldoxime was of some interest. In early work on the tracer scale (80), ên Co** carrier was used with sali- cyleldoxime and ehloroform extractions of PuflT) tracer, par titions as high as 80 per cent were obtained. The experimental results were erratic, however, and were difficult to duplicate. £ven the carrier seesied to have some effect. However, if true inner complex formation occurs, and if all of the inner ooaplex dissolves in the organic solvent, as is the case with cop per and cobalt salicylaldoxiiaes, the erratic results cannot be explained by the carrying properties of the precipitates. Table 15 SztraGtion of Plutoniua lâaer with Salloylaldoxime and Chlorofona from Aoetate-buffered Solatlona

CJarrier i\mount Per Cent in SO ml of Pu pH Solution Reagent £xtraoted Conditions

3.12 to 6.66 None 15 mg 14.1 No NB^OHCl present 3.04 to 6.58 1 IBg Ou*"*" 15 mg 10.0 m NH3OHOI present 3• 0@ to &.5S 2 mg Go*'*' 15 mg 0.7 So NH3OHGI present 6.11 33.5 6.58 11.3 5.54 5 Qtg Co*"*" 40 mg 64.6 m NH^OHCl present 5.65 52.5 5.73 60.7 6.07 Z lag CO**- 20 mg 63.3 Solution 0.1 H in HH3OHCI. -92-

From the results of the tracer experiments, some faetors indioated that the salloylaldoxime might hare oofflplexed Pu(III) instead of Pu(I7). The greatest ex tractions were obtained when hydroxylamine was present, or when a fairly large amount of cobalt livas used as carrier. Hydroxylamine is known to produce Pu{XII) and certain oobaltous oomplexes are known to be among the most powerful reducing agents known (89)» loweTer, with Ptt(III) on the micro scale {See Table 14) there was no evidence for complex formation with salicylaldoxime. So far, no explanation for these discrepancies has been eTOlved. -93-

711 THi. USE OF OiiQANIC REAGM1B TOR THE DECOKTAKINATION MB PirrtlFlGATION OF PLUTONIUM

7«I lAtroduotloa

After several reagents had been found to form ohelate oompounds with Pu{III) and PuCIV) efforts were directed to ward the detenaination of the speolfioity or seleotivity of the reagents. Some of the elements whose behaviors were stu died were those considered in the purifioation of plutoniom; others were the fission produots whioh had to be removed some time during the proeessing of plutonium from the uranyl ni trate disBolver solution to the final fona in which plutonim. was desired. During the eotirse of the investigations, m.phmiB on various aspeets of the deoontamination and puri fication problems was changed. Therefore, some of the speoifity experiments described below are rather limited in their scope* intended studies were conducted only with the reagents whioh exhibited considerable complexing activity with plutonim» For Pu(I?), the best reagents studied in the MieM laboratories were disalioylalethylenodlimine and disal (di- ^,S-dihydroxy-6 or 6-tert« -butylbenzal7''®^^y3.enedilalne)» 1?he best reagents for Pu{III) were benzohydroxaiaio acid. -94-

7.2 Experimental

The behavior of various oetions vdth disal

This reagent effects greater than 95 per cent oomplex- isg of Pu(IT) and the pH range 8,75 to S,98, The behavior of this reagent with various cations was examined by extraotions of traoer amounts and of milligram amounts of ions from aoetate*buffered solutions (0.1 to 1.0 M in total acetate) by disal and <^loroform. Extractions were made either in sep* aratory funnels or in an extinction apparatus (somewhat lar ger than that shown in Figure 2) in which agitation of phases was produced by a gas stream. In all oases the volume of ohlorofom was equal to that of the aqueous phase, ^dioactive 33 year was used to determine the dis tribution of cesium in disal extraotions. Activities were measured by checking the radioactivity of CsClO^ preoipitates tr

behavior of barium was determined over a wide pH 14Q range with 12,8 day Ba . The tracer was reoovered from the fractions as BaCOg and the activity determined with the eleotroscope* A solution of activity (24 day was prepared by ether extraction of uranyl nitrate. The buffered solution was extraoted with disal and chloroform, and the tracer re covered by La(OH)g precipitations. For extractions of zirconium, a sample of carrier-free zirconium tracer was obtained fron Clinton pile material. Absorption measureirients showed that the activity was ut least 9Q per cent pure 68 day Zr®®. The tracer was recovered by FetOH)^ precipitations and the activities determined witli the electroscope. The behavior of disal with tracer amounts of various aatione is stosmarized in Table 16* 96-

T&ble 16 The Ixtraotion of Tarious Hadioactiv® lleoidiits b7 0isal and Ohloroform from Aoetate Sclutions

pH of A^aeous Per Oent Activity lOB Solutions Extracted by Chloroform cs* 5.88 0.02 4.e8 0.02 Ba** 1.93 0.S4 S.35 1S.5 4, OS IS.5 4.80 17.9 5.12 22.6 5.60 21.5

Se***", t*** S.B8 0.09 3.85 0.37 S.50 12.6 3.08 14.7 3.&£ 25.0 4.53 21.4 5.40 51.2 6. 20 61.0 ZstQ** 1.20 12.2 s.oo 60.3 S.90 83.3 3.96 81.2 4.70 82.6 5.S5 85.4

The data insiioate that ^od separatioas of Pu(IY) from alkali metals aad rare earths are possible, bat that direct separatioas of PulIT) from barium, thoritm aad zirooaium are muoh leas effeetive* -97-

The behavior of iron with disal was studied using milli gram amounts of iron, rhe F©*** solution was prepared froa 99,8 per oent pure ii*on wire. ISxtractions were made from SO lal of aqueous solution, 1 M in acetate, oontainlag 2.50 ag of fe***", la ©aoh eattraotion 40 mg of disal was used with 20 al of ehlorofora. Iron in the fractions was determined speo* trophotometrieally as the ferrous orthophenanthxoline ocsaplez ion. Hie results of several experiments are listed in ^t^ble 17,

Table 17 The Extraction of IroB by Disal and Chloroform from Acetate Solutions

pH of Aqueous Per Oent Iron Ex Solution tracted by Chlorofom

S.6S 7.2 3.13 2.8 4.0S 23.0 4.78 46.4 5.S4 53.8 6.08 41.0

Cursory experiments showed that the iron in the disal oomplex ^as in the ferrous state. In these experiments the ohlorofom solutions were shaken v#ith 1.5 M HGl to destroy the complex. The ferrous iron in the HCl solution was de- -9C- teimiaed speotrophotoiaetrioally as the orthophenanthrolin® oomplex ion. From 72.6 to 88iO per cent of the iroa ia tb® HCl solution was present in the ferrous state, Tery stable emulsions were foraed in disal-ohloroform ©xtraotiona of 5 to 10 mg 'iuantities of and 2rO**' ions, aakiag irupossible aoourate eetiaatee of distribu tion of suGh quantities of material.

Attaaipts to adapt disal extractions to the separation of Plutonium from ztroonium and thorima

Both ziroonium and thorium are quite effeotively com* plexed by disal in the pH range in whioh Pu(I?) is ocmplexed. fh© follo\?ing experiments iwore designed to determine whether there are any eonditions with v,'hioh practical fractionations oan b@ obtained. Zirconium, fhe stability of the Zr(I^)-disal complex toward dilute HHOg v/as eoiapared ^rtith that of the Pu(I?)- disal ccmplex in the folio-wing experiment, Disal extractions were performed from acetate-buffered solutions (pH 4,3 to 4.6) of zirconium tracer in the usual way, the chloroform solutions from these extractions were shaken >-ith equal volumes of dilate HSOs and the chloroform and EKOg phases were analyzed for ziroonium activity. 99

Table 16 Stability of Dlsal Ck>Bpl«ces Toward Dilate Efitrio Aoid

Per Cent Aetivlty ResaoTed from ^rmality of Nitric Aoift Ohlorofora by Qae Aeid Wasli 2P PU

0.05 62.0 83.7 0.10 63.8 O.SO 51.6 09.2 0.40 52.2 99.6 0.60 63.? 0.80 93.5 99.5

ohloroform solutions of the 2;r {i:?)>disel ooMplex were agitated witk equal volumes of 1.0 H aqueous IF solatioa, ?5.5 to 80,6 per eeat of tbe ziroonium aotivity was trans ferred from tlie chloroform to the aqueous phase. fhe stability of the Zr{IT)«di8al ooaplex toward oxalic aeid was determined by agitating chloroform solutions of ae(IT)-disal oomplex with equal volumes of 0.04 M oxalic aoid. From f5.4 to 87.0 per cent of the zirconium activity was trans ferred from the chlorofora solution into the oxalic acid. ^ireonium and thoriim. Because disal is a quadrideatate reagent, it would not be expected to coffiplex FuClIl). There fore, there was a possibility that plutonium could be separated froa and fh(IT) by treating a chlorofoxm solution of 100-

th© disal oomplexea with a reducing solution. tJsed in sueh experiments were hydroxylaapunonium and hydraziniusa chloride and solutions of sodium bisulfite. In all experiments the disal reagent was attaoked by the reducing solutions, and no fractionations were obtained.

7«g«3 fhe behavior of various cations with hydroxaiaio aeids

Benaohydroxamic acid. Benzohydroxamio acid has been found to effect some complexing of Pu(IV) tracer (See Table 2), but the complexing of PuClII) is more nearly complete {See fable 11), It was of interest to determine the behavior of certain other elements with benzohydroxamio acid in the range in which greater than 90 per cent complexing of Ptt{III) takes place. In qualitative teats of the behavior of various cations with benzohydroxamio acid, S to 5 mg of the cation -mre added to 3 ml of 1.0 M sodiom acetate solution* fhe organic reagent was added to this solution and the pH adjusted by addition of acetic aiid or a^monii^ hydroxide. Precipitation and ex traction behavior with chloroform were observed as sussmarized in Table 19* Table 19

i^ualltatiYe Behavior of Various Cations with Benzohydroxaaio Aoid aad Chioroform {Aoetat® Buffer)

Preeipitats Solubility Cation pB in HgO in C«Gl3 Hemarks

6 HOutt Slight turbidity in %0. 6 Hone » Ho visible change. oo**- 6 None - BTOTTO water soluble oolor at pH-8, iSA** 7 Browa SI. sol • Btoulsion formed* 6«8 None - Ho visible ohange. AX*** 3 White •> Turbid mulsion foxmed* Mfe** 8 yaint Insol. Qxk** 6-e areen SI. sol. CHOls layer turbid. 3-6 Pale Tallow SI. sol. Itxtraotion better at higher pH.^ Ag^ S-6 Furplish-llaolc Sol. ?•••• 4»S Red Sol. XJQz** 3*4 Hoae Water-soluble briolc-red color. S-6 Hone 6 Hooe > o«r*t. 8 Brotm Sol. 3 mite lasol. ZrO** S<-6 White IQSOL. OaJJ 5 Roue • Slight turbidity in HgO* Ife 6 Hose « La*'*'*' 8 None «• 8 Orange Sol. -108-

Tlie results indicate that benzohydroxamic aeld foms ehelate compounds with so many oations that it oannot be regarded as a seleotiTe reagent. °<*Kaphthohydro3:a3alo aeld. It was Important that the hehavior of this reagent with zireoniua and thoriu® ilSX^) be inTestigated. The tracer scale eztraetlons were done in the asual way, using 68 day za?®® and 24 day {X3Ki) activities* Under the same oonditions under which Fu{IIX} extraotions are nearly complete, SP{I7) and ThilY) extrao- tiOBS are appreciable. Table 20 contains the results of two such experiments.

Table ZO Sxtraction of Tracer Ziroonlm and fhori\ia by -Baphthohy^^roxamie Acid and Ohlorofoxm froa Acetate Solatioaa

pH of Aqueous Per Cent Activity Sx- Element Solution traoted by ChlortJform

^oriaa 3.7S 0.85 4.39 58.2 4.90 6S.5 5.40 n.9 Zirconivoi 4*04 94.5 4^50 92.5 4.9e 86.0 5.40 86.6

The data show the similarity of ThdV) and ZrCiT) to -103-

Pu(III) In their activity toward chelating reagentsi

The hebevior of varioag eatlona with amide oximes

To obtain a qualitative idea of the speoifity or seleetiTity of phenylacetamide oxime, a series of experi ments was performed in which 3 to 5 ag of oation were al lowed to react with phenylacetamide oxime in 3 ml of acetate-huffered solution at pH 5 to 6, Precipitation and Bolufeility properties are compiled in Table 21.

Table 21 Qualitative Behavior of Milligram Amounts of Various Cations with Phenylacetasiide Oxiffle in Acetate Solntions at pH 5 to 6

Solubility of Preci Cation Precipitate pitate in Chloroform

Tellow-orange Soluble Ott*" Olive green Soluble Hi*"* Hone re*'^*- Bed Soluble White Insoluble (^droxide) Xta Hone Oe^**- ?lhite In soluble {Hydroxide) White Insoluble (Hydroxide) Qreeniah-brown Soluble 104-

It oaa be seen that phenylaoetamide oxime is very similar to henzohydroxamio aoid in its reaotions with various eatiohs. For the experiments designed to study the possible separation of Pu{III) from Zr(IT) a solution oontaining both aotivities was obtained from J. A. Ayres. laoh ex* periiiient {Table 22) involved the use of about 43 //g of plutoniuB and about 160,000 oounts per minute of ziTQonixm activity la S ml of acetate-buffered solution. In all of the experiments, phenylaoetamide oxiae, ^hioh is fairly soluble in water, was added to the buffered solution and allowed to react for 2 hours before extraction with chloro form. The extraction step was pr«doded by a reduction step, ^ree different reducing agents were studied to compare the effect of different media on phenylacetamide oxime.

Table 2£ The Extraction of Plutonium(III) and ^irconiumd?) by Phenylacetamide Oxime and Chloroform from Acetate Solutions at pH 5 to 6

BOB Conditions for Re Per Gent Activity ducing Plutonium gxtracted by Ohlorogora Pu(III) 2r(IT)

Saturated SOo solution, 4 hrs. 0.1 0 Z M lSg0a>HQl, HCl, 4 hrs. 91.9 37.0 3 M SsB4.m, HCl, 4 hrs. 64.2 59.2 -105-

Agaln, Zr{17) shows the same persistent similarity to Pu(III) in its reactivity with organic reagents,

7.3 Discussion

None of the reagents discussed above is applicable to the direct extraction of plutoniiaa from dissolver solutions of uranyl nitrate. Disal,c<-naphthohydroxamic acid, and phenylaeetamide oxime for chelate compounds with 'OOg*''' and ions* as well as with many other ions. However, some of the reagents find practical use in many laboratory operations with Plutonium. Disal effects very good separations of plutonium from the rare earths. This reagent will remove tracer amounts of FuClT} from solutions containing large amounts of lanthanum. Since LaFg is a very common carrier for tracer amounts of Pu(III) or Pu(IY), disal extractions make available a versa tile method of recovering plutonium from miscellaneous solu tions. In a typical recovery procedure, LaFg precipitations are made from solutions to concentrate the plutoniuml The fluoride is converted to the sulfate by fuming to dryness with HgSO^, the sulfate taken into solution and buffered to pH 5 to 4 for disal extractions. In this way, plutonium is separated from rare earths and lantheuaum without the oxidation-reduction procedures used earlier in project history. Disal extractions have been used repeatedly in this labora 106-

tory for reoovery purposes and have been found to give con sistently good results. It can be seen from the data in Table 17 that disal extractions are not applicable to the separation of pluto- niun from iron. However, by preceding disal extractions with the fluoride precipitatioh step described above, very good separations from iron are obtain^. The behavior of Th(IV) and 3r(IV} with organic rea gents for plutoniiaa was of especial interest. The solu tion chemistry of these two elements is quite similar In ma:!^ respects to that of plutonium. This is especially true in the adsorption process for extraction and deoontamination of plutonim (81). The principal contaminants in the pro duct solution from this process are Zr(nr), one of the fis sion products, and Th(IT) as the daughter of by -emission. The behavior of Th{I?) and Zr(IV) with disal is remark ably like that of Pu{IV). In extractions with disal and chlo roform, TxiH) is extracted almost as completely as Pu{17) ia comparable pH ranges, and the stability of the Zr{IT)-disal complex toward dilute HKOg Is too similar to that of the Pu(iy)-disal complex to make possible practical separations of from Pu(IT) by appropriate acid re-extraction cycles. The same phenomena are true for oxalic acid re-extractions* The relative efficiency of disal extractions of Th(IY) is 107-

eomewhat less than that for Pu(I7) at different pH»s, hut the differences are not great enough to be of use in sepa rating Pu(I¥) from Th(I7). The eiiailarity of Th{IT) and ardV) to Pu(lT) in re actions with chelating reagents has been noticed by others, and some investigators have used Th{17) as a "stand-in" sub stitute for Pu{rf) in exploratory studies with organic rea gents (91}• A possibility for separating ZrdTT) and Th(I7) from plutoniua arises from the faot that the chelating properties of FuCIII) differ from tiwjse of Pu{IT). Some of the ^-diketones, for example, oomplex Pu{IT) and Zr(I'^)» t)ut not Pu{III) (S2). Thus, a benzene solution of the Pu(IT) and 2r(IV) che lates is treated with a reducing solution, such as hydroxyl- amonium chloride, hydrosulfite or stannous solution, so that the Fu{III) formed is transferred to the aqueous layer, leav ing the ZriXV) chelate in the benzene. Similar reduction procedures are not effective with disal complexes. The dis&l complexes are less stable toward acids than are the diketonates, and will not withstand the acid concentrations necessary for the reduction of Pu{r?) to Pa(III). At pH*e high enough for stable disal complaxas-fc. hydroxylanuBonitim chloride converts the Schiff's base to the corresponding oxime of the aldehyde, and SOg reduces th« diimine to a diamine which has no complexing properties. The hydroxaiaio acids are quite reactive in forming che- 108- late ooapouadfi with many cations (87, 93), so their appli cability to the deoontamiaatlon and purification of plutoniua is rather limited. The behavior of Zr{IV) and Th{IT) with c<-Qaphthohydroxamic acid {See Table ZO) is so similar to that of Pu(III) that separations of the three elîents oaanot be achieved in a single operation, Oxidation-reduc- tioa cycles, in combination with extraction operations, are of no value, either, because <<-naphthohydroxamie acid also eomplexes Pu(IV"). The amide oximes are quite similar to hydroxaaie acids in their reactions with many cations (93), and their appli cability to general problems of decontamination and purifi cation of Plutonium is limited. Like 'K-naphthohydroxamie acid, phenylacetamide oxime does not effectively separate Pu{III) from 2ar(IT). With hydroxylanaaonium chloride as re- ductant, ho'iêver, phenylacetamide oxime extractions riKaoved 37 per oent of ZrÎV) compared with 86 per cent extraction of 2r{Iir) byt^-naphthohydroxamic acid under similar condi tions. Phenylacetamide oxime in the presence of SGg was in active toward both Pu(III) and Zr{17), probably because of reduction of the oxime to an amine. Phenylacetamide oxime has some activity for Pu(IY), as well as for Pu{III) so oxidation-r@ductioQ cycles would be of no value in plutonium- airconium separations. -109

VIII SUMMARY

1. A brief dlsousslon of the chemioal properties of plutonium hae been presented., with emphasis on the solation ehemistry of plutoniua. The general chesaical propertiea of plutonium aad other transuranium elements, along with the absorption speotra of their aqueous solutions, indicate that they may be members of a transition group in whioh the 5f subshell is being filled. B, The various faotors involTed in the formation of ohelate compounds have been reviewed. 3, Esploratory experiments with organic reagents and tracer plutonium have been shown to be satisfactory for quali tative tests for chelating activity of organic reagents for plutonium. In most of the tracer scale experiments, the concentrations of plutonium were of the order 10••10 to 10"—8 M. Except in the case of salicylaldoxime, positive results ob tained from exploratory tracer scale experiments were veri fied in experiments in whioh amounts of plutonium from 100 to 3 mg were used. 4. In exploratory experiments with Pu(I7) tracer, 45 different bidentate reagents were examined for chelating activity. All of the reagents examined were loiown to form chelate compounds with certain metal ions. Chloroform was used as the solvent in most of the experiments, end, using -110

a Toluae of chloroform equal to that of th« buffered aque- oijs solution, extraotabllities greater than 10 per oent were taken as eTidenoe of complex formation, Bidentate reagents which foimcHl inner oomplexes with Pu(IV) include quinalizarin, o^-hydroxyhenzophenone, benzoylacetone, dibenzoylme^ thane, trifluoroaoetylaeetone, benzohydroxamie acid, phenylaoetamide oxime, isonitrosoaoetophenone, salieylal-^^-hy- droxyphenyl^-imine, £-hydroxyacetophenone eemioarbazide, and e-^'hydroxyquinoline. 5. Systematic studies of the behavior of 26 quadridentate reagents with Pu{IT) tracer were made. The most ac- tiT# quadridentate chelating reagents are the Schiff's bases prepared trm. ethylenediaiaine and Tarious derivatives of salioylaldehyde. Hineteen different derivatives of disali- oylalethylen€>diiffiine were compared in their chelating ac tivity with Pa{IT) tracer. The Sohiff*s base which effected most complete chelating activity with Pu(I7) over the great est range of hydrogen ion concentration was dl-j^,3-dihy- 4roiy-8 or t-tart.-bntyl-benzalj-ethylenaailmlne. refeiT«4 to in this thesis as "disal", 6. Exhaustive attempts to prove the position of the tert>-butyl radical in disal were unsuccessful. Positive proof of structure depended on the oxidation of the tert«- butyl radical to a carboxyl group, and tertiary alkyl radi cals are peculiarly resistant to oxidation. 7. Fourteen different solvents, including hydrocarbons. -Ill- ahlorinated hydrooarbons, alcohols, ethers, esters, and ke- toaes were found to be satisfactory for extraetioas of the Pu{IV)-dlsal complex, 8. The deterrainatioa of the composition of the Pu(IT)- disal complex gare definite ©Tidenoe that the ooordiaatioa number of Fu{I?) is eight. Orbitals which might be inrolTed in S-ooordinate complexes vvere discussed. 9. I'he spectrophotoraetrie behavior of the Pu{IT)-fer- rOB complex was studied, and It ivas suggested that a method for the speetrophotometric estimation of iron In plutoniixa, with ferron as reagent, is possible. 10. Twenty-seven different bidentate reagents wer« examined for their chelating activity with Rea gents v/hieh form chelate compotmds with include 8- hydroxy^ quinoline, benzohydroxamic acid, a^-nitrobenzohy- droxaxaio acid, o-hydroxybenzohydroxamlo acid, ^-naphthohydroxamic acid, phenylacetohydroxqpciic acid, n-valerohy- droxaaie acid, phenylaeetasiide oxime aM nf'Veleraraide oxime. 11. The composition of Pu{III)-

IX LITmATDRE OITED

1. HeMillaa, S, M., and p. H. Abelson, Phys. Rev., 5?, 1185 (1940). —" — S. Seaborg, G. T., S. M. Moiailaa, A. G. Wahl, and f. W. Seimedy, Phye. Rev.« 69, 366, 367 (1946), 3, Babr, H., and A. Wheeler, Piiys. Rev.. 420 (1939). 4# ^Paraer, L. A., Phys. Rev.. 57. 950 (1940). 5. AadersoQ. H. L.• S. fezmi. and fi. B. Eansteln. Phys. Rev.. 797 (1^9). 6. Smjth, 1. X)., "AtOdlG iner^ for Military Pmrpoeea'*, Princeton milversitj Prees, Prinoetoa, 1945. 7. Seaborg, 0. T., Qhaa. Sag. Hews. 23, E190 (1945). 8. !£&omas, 0. A., and J. 0. Wairaer, Chmlstry, PurifJL- eatioB and ISatallargy of Plutoaii:i&'', Manhattan Dietriet, 1945. 9. "Plutoaim Projeot Reoord**, Dlvisioh 27 of the Manhattan Projeot Teohnioal Series, 1946, Tolxmes XITA and XITB. 10. Hindman, J. C., Metallurgical Projeot Report CH-1567, p. 2, May 1, 1944. 11. Waters, J, I., Metallurgioal Projeot Report GK-1371, p. 10, liaroh 1, 1944. 12. quill, L, L., Ofaem. Rev.. 98 (1938). 13. Tost, IL, H. BiU98ell and C. S. Garner, **Advanoed Inor- ganio Ohemistrj**, privately printed in the U. S. A., 1944, Oh. 13. 14. Beaborg, 0. T.. Gh^. Bag. Heiwa. 24, 899 (1946). 15. liorgan, G. T., and H. D. K. Drew, ghem. Soe.. 117. 1456 (19S0). 16. liTohnson, JT. R., in *0rgaalo Ghoalstry", edited by Henry Gilmaa, John Wiley and Sons, Mew Tork, 1943, p. 1868. -114-

17. Diehl, H., Oheia. Rev.. 39 (1937). 18. Mellon, H. 0., "Golorimetry for ChemistB", G. F. Smith Chemioal Co., Columbus, Ohio, 1945, p. 11. 19. (a) Mellan, I., "Orgaaio Reagents in Inorganic Analy sis'*, Blakiston Company, Philadelphia, 1941. (b) Yoe, 3". H., and L. A. Sarrer, "Organic Analytical Reagents", John Wiley and Sons, New York, 1941. (c) Feigl, F., "Specific and Special Reactions for Use in quantitative Analysis", (translated by R. E. Oesper), Elsevier Publishing Company, Hew York, 1940. (d) Prodinger, W., "Organic Reagents Used in Quanti tative Inorganic Analysis", (translated by S. Holmes), Elsevier Publishing Coapany, Sew York, 1940. (e) Hopkins and Williams, "Organic Reagents for Metals", Hopkins and Williams, Ltd., London, 1944. 20. Sidgwick, H. Y., "Electronic Theory of Valency", Ox ford University Press, Oxford, 1929, 21. Sidgwick, S. v., J. Ghem. Soc.. 1941. 433. 22. Pfeiffer, P., in iC. Freudenberg*s "Stereochemie", Franz Beuticke, Leipzig, 1933, pp. 1200-1377. 23. Pfeiffer, P., Angew. Chem., 53, 93 (1940). 24. Pfeiffer, P., J. prakt. Cheia.. 162. 279 (1943). 25. Pauling, L., "The Kature of the Chemical Bond", 2nd ed,, Cornell University Press, Ithaca, N. Y., 1942, pp. 92-123. 20. Kimball, G. 1., J. CheH. Phys.. 8, 188 (1940). 27. Ingold, C. K., Chem. Rev.. ]^, 225 (1934). 28. Orlemahtt, E. F., Metallurgical Project MemoltanduHi, MUG-GTS-8©8^ July 17, 1944, p. 8. 29. Spac.;. G» , and P. Toichescu, Z. anorg. allgem. Chem., 22®, 273 (19S6). -115

30. Calvin, M., aM K. W. Wilson, 3". Am. Chem. Soo.. 2003 (1945). 31. Bailar, J. 0., Qhem. ReT.. 65 (1938). 3S. Gkjpley, M. J., L. S. Foster, and J. 0. Bailar, Chem. ReT.. S27 (1942). S3, Selwood, P. w., "Magnatoohamistry", Intersoienoe Pub lishers, Hew Yorlc, 1943, Oh. 6. 34. Tsohugaeff, L., J. prakt. Chem.. 75. 153 (1907). 35. Werner, A., Ber.. 41. 1062 (1908).

36. Pfeiffer, P., 1. Breith, S. Lubbe, and T. Tsumaki, Ann., 503. 84 (1933). 37. Dent, C. S,, and R. P. Linstead, J. Chem. Soc., 1934, 1027. 38. Blanchard, A. A., J. Chem. Education. 20, 459 (1943). 39. Langham, W. H., Metallurgical Project Report GE»2254, ao date, p. 9. 40. Metallurgical Project Handbook CL-697, 2nd ed., 1945, Oh. 711. 41. Simpson, J. A., Metallurgical Project Report CP-1817, May 8, 1945. 42. Bennstedt, H., and J*. Ziimermann, Ber., 21, 1553 (1888). 43. llkins, M., and L. Hunter, Chem. Soe«, 1935, 1598. 44. Blatt, A. H., ''Organic Syntheses**, Collecti-re Volume II, John Wiley and Sons, Hew York, 1943, p. 67. 45. Delephine, M., ]^11. soo. ohim., /4/, 3, 641 (1908). 46. Tiemaan, F., and P. l&uger, Ber.. 17. 1(^5 (1884). 47. Liggett, L. M., ••The Synthesis of o-Hydroxyaldehydes", Ph. D. Thesis, Iowa State Co liege,""Ames, 1943. 48. f, «r. G*, 547. 49. Combes, A., and C. Combes, Bull. soo. ohim., /3/, 7, 779 (1892). " 50. Meyer, V., Ber., 10. 2075 (1877). -116-

51. MoConnan, , and M. 1. Maroles, JT. 01iem« Soo., 91. 194 (1907). — 52* Bamberger, 1., and W. Pemsel, Bar.. 56. 62 (1903). 53. Hollefsoa, G, K., and H. W. Dodgon, Metallurgical Pro ject Heport CK-1148, December 13, 1943. 54. Urbain, M. C., Bull, soc. ohim.. /3/ 15, 347 (1896). 55. BiltZj W., and J, A. Clinoii, ^ anorg. Ch«a., 40, 218

56. Job, A., and P. Golssedet, Gompt. rend., 157, 50 (1913). 57. Dixon, J. S., Metallurgioal Project Report GE:-2C®9, September 1, 1944, p. 16. 58. Galrin, H., Metallurgical Project Report GH-2486, De- o^ber 1, 1944* 59. Haedel, W. J., A. J". Margolis, and 3". Seidlet, Ifetal- lurgical Project Report GH-S496, January 1, 1945, p. 33. 60. Grandall, H. W,, J, R. Thomas, and J, 0. Reid, Metal lurgical Project Report CN-2657, Janiuiry 15, 1945. 61. Winner, B. M., Metallurgical Project Report CN-2887, April 28, 1945, p. 7. 62. Grandall, H. ?/., and J. R. Thomas, Metallurgical Pro ject Heport ON-2999, May 9, 1945. 63. Gowaa, G. A., and M. Goldsmith, Metallurgical Project Report GK-1562, April 1, 1944, p. 5. 64. Langham, W. H., Q. A. Cowan, S, Wexler, H, S, Dunham, and H. A. Potratz, Metallurgical Project Report C5- 2254, no date, p. 9. 65. Thompson, S. G., Metallurgical Project Report CII-328, October 31, 1942, p. 10. 66. Willard, J. E., and E. E. Turk, Metallurgical Project Report GH-282, September 30, 1942, p. 7. 67. Patton, R. L,, Metallurgical Project Report GK-1587, April 1, 1944, p. 15. 68. Dixon, J. S., Metallurgical Project Heport GK-1763, duly 1, 1944, p. 4. -117-

69. Combes, A., and C. Combes, Goiapt. read., 108, 1252 (1889). 70. Pfeiffer, P., and H. Pfitzner, J", prakt. Chaia., 145, S43 (1936). 71. Pfeiffer, P., T. Hesse, H. Jfi^siner, W. Soholl, and H. Tiiielert, J. prakt. Chem.. 149. 217 (1937). 78. Dixon, J. S., Metalltirgioal ^rojeot Report OiC-1763, July 1, 1944, p. 4. 73. XRiffield, B. B., and M. OalTin, 3". Ara. Gheci. Soo., 68, 557 (1946). 74. fieoht, F., and W. Reich-Rohrwi#?, i&)nat8h., 53-54, 596 (1929). 75. ?rer6, 5*. 3"., 3^ 55, 436S (1^53). 76. Marahi, L. E., W. C. Fernelius, and J". P. MoReynolds, Ibid.. 65. 329 (1943). 77. Hoard, 3". L., and H. H. Nordsieok, ibid., 61. 2853 (1939). — 78. jferohl, I,. S., and 3". P. MoReynolds, ibid., 65, 333 (1943). 79. Pfeiffer, P., H. Thielert, and H. Qlaser, 3". prakt, Ohem.., 152. 145 (1939). — 80. Clark, R. S., Metallurgioal Project Report GN-2519, August 1, 1944, p. 54. 81. Ayres, J. A. , Metallurgioal Project Report GII-2718, May 11, 1945. 82. Yoe, J., 3". Am. Ghem. Soe., 54. 4139 (1932). 83. Yoe. J., and R. T. Hall, ibid.. 872 (1937). 84. Swank. H. W.. and M. G. Mellon. Ind. Eng. Ohem.. Anal. M., 9, 406 (1937). 85. Hindman, I. G., and D. Ames, Metallurgical Project Re port CS9-2289, November 1, 1944. 86. Tale, H. L., qhem. Rev.. 209 (1943). 87. Werner, A., Ber.. 41, 1062 (1908). 118

88. Wolter, F. 5"., and H. D. Bro-wn, Metallurgical Pro ject Report CN-1495, April 10, 1945, p. 17. 89. Lattaer, W. M., '•Oxidation Potentials", Prentieo- Hall, Inc., New York, 1938, p. 201. 90. Wolter, P. J. , H. D. Brown, J. M. Wright and R. R. Raeuohie, Metallurgical Project Report ON-1783, August 10, 1944, pp. 13-19.

91. Cowan, G. A., R. IS. Fryxell, and H. A, Potratz, Metal lurgical Project Report GE-993, October 9, 1943. 92. Orandall, H. ??., and J. R. Thomas, lletallurgioal Pro ject Report CH->2999, May 9, 1945. 93. Lossen, W., Ann.. 161, 347 (1872). 119'

X AClCNOWLEDffillNfS

The vsriter expresses his iadebtedaess to those w>» have helped make these studies possible. Dr. F. H, Spedding, as Projeet Direotor, suggeeted the problem, and has given overall superviBion and en- oo'irageraent during the development of the probl^. Dr. A. F, Toigt was in iiimediate charge of the writ er's researoh, and his suggestion® and constant cooperation are deeply appreciated. Br, Harvey Diehl, as consultant for the Project, gave many worthwhile suggestions, and furnished many of the or ganic cofflpounde used in the work. H. B. Brov^n assisted in much of the v»rk with tracer and aiero amounts of plutoniiim, and 3". M, Wright did many of the exploratory tracer scale experiments. R. F. Haauchle assisted in the preparation of several of the organic rea gents examined. Other members of the Project have also given their com plete cooperation, without which many of the studies would have been much more difficult. Sincere appreciation ^joes to Dr. I. B. Johns, who di rected the writer*s graduate studies before they \'wre in terrupted by the war. The researoh described in this thesis was sponsored -120-

by the Corps of Sagineers, Manbattan Distriot, UGite4 States iiraiy, and the funds for Its support came from Ooatraot W-7405-€ng-82 between the ManliattaQ Distriot aad Iowa State College.